Continual learning for robotics: Definition, framework, learning strategies, opportunities and challenges
•State of the art on continual / lifelong learning and its implications for robotics.•Proposal of a framework to present Continual Learning algorithms.•Summary of existing metrics for continual learning.•Presenting robotics applications for continual learning, current challenges and opportunities. C...
Gespeichert in:
| Veröffentlicht in: | Information fusion Jg. 58; S. 52 - 68 |
|---|---|
| Hauptverfasser: | , , , , , |
| Format: | Journal Article |
| Sprache: | Englisch |
| Veröffentlicht: |
Elsevier B.V
01.06.2020
Elsevier |
| Schlagworte: | |
| ISSN: | 1566-2535, 1872-6305 |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| Abstract | •State of the art on continual / lifelong learning and its implications for robotics.•Proposal of a framework to present Continual Learning algorithms.•Summary of existing metrics for continual learning.•Presenting robotics applications for continual learning, current challenges and opportunities.
Continual learning (CL) is a particular machine learning paradigm where the data distribution and learning objective change through time, or where all the training data and objective criteria are never available at once. The evolution of the learning process is modeled by a sequence of learning experiences where the goal is to be able to learn new skills all along the sequence without forgetting what has been previously learned. CL can be seen as an online learning where knowledge fusion needs to take place in order to learn from streams of data presented sequentially in time. Continual learning also aims at the same time at optimizing the memory, the computation power and the speed during the learning process. An important challenge for machine learning is not necessarily finding solutions that work in the real world but rather finding stable algorithms that can learn in real world. Hence, the ideal approach would be tackling the real world in a embodied platform: an autonomous agent. Continual learning would then be effective in an autonomous agent or robot, which would learn autonomously through time about the external world, and incrementally develop a set of complex skills and knowledge.Robotic agents have to learn to adapt and interact with their environment using a continuous stream of observations. Some recent approaches aim at tackling continual learning for robotics, but most recent papers on continual learning only experiment approaches in simulation or with static datasets. Unfortunately, the evaluation of those algorithms does not provide insights on whether their solutions may help continual learning in the context of robotics. This paper aims at reviewing the existing state of the art of continual learning, summarizing existing benchmarks and metrics, and proposing a framework for presenting and evaluating both robotics and non robotics approaches in a way that makes transfer between both fields easier. We put light on continual learning in the context of robotics to create connections between fields and normalize approaches. |
|---|---|
| AbstractList | Continual learning (CL) is a particular machine learning paradigm where the data distribution and learning objective changes through time, or where all the training data and objective criteria are never available at once. The evolution of the learning process is modeled by a sequence of learning experiences where the goal is to be able to learn new skills all along the sequence without forgetting what has been previously learned. Continual learning also aims at the same time at optimizing the memory, the computation power and the speed during the learning process. An important challenge for machine learning is not necessarily finding solutions that work in the real world but rather finding stable algorithms that can learn in real world. Hence, the ideal approach would be tackling the real world in a embodied platform: an autonomous agent. Continual learning would then be effective in an autonomous agent or robot, which would learn autonomously through time about the external world, and incrementally develop a set of complex skills and knowledge. Robotic agents have to learn to adapt and interact with their environment using a continuous stream of observations. Some recent approaches aim at tackling continual learning for robotics, but most recent papers on continual learning only experiment approaches in simulation or with static datasets. Unfortunately, the evaluation of those algorithms does not provide insights on whether their solutions may help continual learning in the context of robotics. This paper aims at reviewing the existing state of the art of continual learning, summarizing existing benchmarks and metrics, and proposing a framework for presenting and evaluating both robotics and non robotics approaches in a way that makes transfer between both fields easier. •State of the art on continual / lifelong learning and its implications for robotics.•Proposal of a framework to present Continual Learning algorithms.•Summary of existing metrics for continual learning.•Presenting robotics applications for continual learning, current challenges and opportunities. Continual learning (CL) is a particular machine learning paradigm where the data distribution and learning objective change through time, or where all the training data and objective criteria are never available at once. The evolution of the learning process is modeled by a sequence of learning experiences where the goal is to be able to learn new skills all along the sequence without forgetting what has been previously learned. CL can be seen as an online learning where knowledge fusion needs to take place in order to learn from streams of data presented sequentially in time. Continual learning also aims at the same time at optimizing the memory, the computation power and the speed during the learning process. An important challenge for machine learning is not necessarily finding solutions that work in the real world but rather finding stable algorithms that can learn in real world. Hence, the ideal approach would be tackling the real world in a embodied platform: an autonomous agent. Continual learning would then be effective in an autonomous agent or robot, which would learn autonomously through time about the external world, and incrementally develop a set of complex skills and knowledge.Robotic agents have to learn to adapt and interact with their environment using a continuous stream of observations. Some recent approaches aim at tackling continual learning for robotics, but most recent papers on continual learning only experiment approaches in simulation or with static datasets. Unfortunately, the evaluation of those algorithms does not provide insights on whether their solutions may help continual learning in the context of robotics. This paper aims at reviewing the existing state of the art of continual learning, summarizing existing benchmarks and metrics, and proposing a framework for presenting and evaluating both robotics and non robotics approaches in a way that makes transfer between both fields easier. We put light on continual learning in the context of robotics to create connections between fields and normalize approaches. |
| Author | Filliat, David Stoian, Andrei Maltoni, Davide Lomonaco, Vincenzo Díaz-Rodríguez, Natalia Lesort, Timothée |
| Author_xml | – sequence: 1 givenname: Timothée surname: Lesort fullname: Lesort, Timothée organization: Flowers Team (ENSTA Paris, Institute Polytechnique de Paris & INRIA) France – sequence: 2 givenname: Vincenzo surname: Lomonaco fullname: Lomonaco, Vincenzo organization: Department of Computer Science and Engineering - University of Bologna – sequence: 3 givenname: Andrei surname: Stoian fullname: Stoian, Andrei organization: Thales, Theresis Laboratory – sequence: 4 givenname: Davide surname: Maltoni fullname: Maltoni, Davide organization: Department of Computer Science and Engineering - University of Bologna – sequence: 5 givenname: David surname: Filliat fullname: Filliat, David organization: Flowers Team (ENSTA Paris, Institute Polytechnique de Paris & INRIA) France – sequence: 6 givenname: Natalia surname: Díaz-Rodríguez fullname: Díaz-Rodríguez, Natalia email: diaz.rodriguez.natalia@gmail.com organization: Flowers Team (ENSTA Paris, Institute Polytechnique de Paris & INRIA) France |
| BackLink | https://hal.science/hal-02381343$$DView record in HAL |
| BookMark | eNqFkE9rGzEQxUVJIHGSb5DDXgverUar_edDwbhtEjDkkpyFVh65464lI8kp-fbdxSmBHprTDMN7b3i_GTtz3iFjt8AL4FB_2RXkrD3GQnDoChAF5_ITu4S2EXld8ups3Ku6zkVVVhdsFuOOc2h4CZeMVt4lckc9ZAPq4MhtM-tDFnzvE5m4yL6hJUeJvJtnNug9_vbh1_xdHVPQCbeEcZ75w8GHdJzkGDPtNpn5qYcB3RbjNTu3eoh48zav2POP70-r-3z9ePewWq5zIyuR8rbvDPAerW1bgb1pq04aI80GQcqqldoCNkY3stYCEUxtuJSdFXXXaIS2L6_Y51Pu-FkdAu11eFVek7pfrtV046JsoZTlC4zaxUlrgo8xoFWGkp66jqVoUMDVBFjt1AmwmgArEGoEPJrlP-a_3z6wfT3ZcITwQhhUNITO4IYCmqQ2nv4f8Acp4pwK |
| CitedBy_id | crossref_primary_10_1371_journal_pone_0331139 crossref_primary_10_3390_math13182988 crossref_primary_10_1063_5_0186054 crossref_primary_10_1016_j_anucene_2022_109323 crossref_primary_10_1109_ACCESS_2021_3059251 crossref_primary_10_1109_LRA_2022_3157369 crossref_primary_10_1109_TPAMI_2025_3568886 crossref_primary_10_1109_JPROC_2025_3581995 crossref_primary_10_1109_TVT_2024_3398661 crossref_primary_10_1088_1361_6501_ad25e3 crossref_primary_10_1016_j_inffus_2023_101896 crossref_primary_10_3389_fnbot_2021_658280 crossref_primary_10_1016_j_neucom_2025_129510 crossref_primary_10_1145_3451883 crossref_primary_10_1145_3711121 crossref_primary_10_3390_pr13072197 crossref_primary_10_1016_j_future_2024_04_046 crossref_primary_10_1016_j_cviu_2021_103167 crossref_primary_10_3390_technologies9040086 crossref_primary_10_1109_TITS_2022_3227122 crossref_primary_10_1109_TAFFC_2022_3181033 crossref_primary_10_1109_TAES_2025_3544593 crossref_primary_10_1016_j_inffus_2021_09_022 crossref_primary_10_1016_j_engappai_2022_105549 crossref_primary_10_3390_fi16110413 crossref_primary_10_1016_j_geits_2023_100103 crossref_primary_10_1109_MIS_2024_3479469 crossref_primary_10_1016_j_compeleceng_2024_109220 crossref_primary_10_1109_TPAMI_2024_3446949 crossref_primary_10_1049_cvi2_70013 crossref_primary_10_1109_TKDE_2024_3447123 crossref_primary_10_1016_j_neucom_2025_131014 crossref_primary_10_3390_app112412078 crossref_primary_10_1109_TIP_2021_3106799 crossref_primary_10_1016_j_patcog_2023_110238 crossref_primary_10_1016_j_cosrev_2023_100548 crossref_primary_10_1016_j_wasman_2025_02_051 crossref_primary_10_15803_ijnc_14_2_123 crossref_primary_10_1109_TAC_2025_3536839 crossref_primary_10_1016_j_engappai_2024_109143 crossref_primary_10_1016_j_compag_2021_106608 crossref_primary_10_1109_TIM_2023_3265118 crossref_primary_10_1016_j_neunet_2023_01_014 crossref_primary_10_3389_frobt_2021_683066 crossref_primary_10_1016_j_inffus_2021_10_007 crossref_primary_10_3390_jimaging7090180 crossref_primary_10_1109_TKDE_2025_3553162 crossref_primary_10_1007_s10462_024_10853_9 crossref_primary_10_3389_frai_2022_829842 crossref_primary_10_1002_jocb_70051 crossref_primary_10_1109_TNNLS_2023_3347453 crossref_primary_10_1111_brv_13054 crossref_primary_10_1016_j_robot_2022_104294 crossref_primary_10_52468_2542_1514_2021_5_1__185_201 crossref_primary_10_1109_MCI_2021_3129959 crossref_primary_10_1016_j_jmsy_2025_08_018 crossref_primary_10_1109_JIOT_2025_3534242 crossref_primary_10_1007_s00170_024_14508_5 crossref_primary_10_3390_healthcare12020155 crossref_primary_10_1016_j_neunet_2021_07_021 crossref_primary_10_1007_s13042_025_02663_4 crossref_primary_10_1016_j_apenergy_2025_125458 crossref_primary_10_1109_LRA_2021_3104334 crossref_primary_10_1109_JSEN_2024_3486021 crossref_primary_10_1007_s13132_024_01879_z crossref_primary_10_3389_fnins_2023_1196796 crossref_primary_10_1109_TRO_2023_3306615 crossref_primary_10_1016_j_engappai_2023_105927 crossref_primary_10_1109_TNNLS_2024_3446171 crossref_primary_10_1016_j_cogsys_2025_101377 crossref_primary_10_1145_3712306 crossref_primary_10_1007_s00521_021_06154_9 crossref_primary_10_1016_j_neucom_2021_10_021 crossref_primary_10_1016_j_compbiomed_2023_107412 crossref_primary_10_3390_electronics12040866 crossref_primary_10_3389_fdata_2020_577974 crossref_primary_10_1016_j_inffus_2022_07_024 crossref_primary_10_1109_JSTARS_2025_3588052 crossref_primary_10_1109_TNNLS_2024_3478334 crossref_primary_10_1038_s41598_023_42482_7 crossref_primary_10_1016_j_neucom_2025_129807 crossref_primary_10_1109_TPAMI_2022_3213473 crossref_primary_10_1007_s43154_021_00068_z crossref_primary_10_3389_fnbot_2020_620378 crossref_primary_10_1080_01691864_2022_2035253 crossref_primary_10_1016_j_knosys_2025_113477 crossref_primary_10_1080_19349637_2025_2454427 crossref_primary_10_1126_sciadv_adu6991 crossref_primary_10_1016_j_robot_2023_104568 crossref_primary_10_1109_LRA_2022_3203812 crossref_primary_10_1109_ACCESS_2025_3561103 crossref_primary_10_1109_TITS_2023_3324983 crossref_primary_10_1109_LRA_2020_3010753 crossref_primary_10_1109_LRA_2024_3418310 crossref_primary_10_3390_s22186881 crossref_primary_10_1109_ACCESS_2024_3510536 crossref_primary_10_1080_01691864_2023_2225232 crossref_primary_10_1007_s11227_023_05241_z crossref_primary_10_3389_frobt_2022_669420 crossref_primary_10_1109_ACCESS_2024_3381967 crossref_primary_10_1016_j_engappai_2025_110908 crossref_primary_10_1088_2632_2153_adeb45 crossref_primary_10_1109_TASE_2025_3569402 crossref_primary_10_1109_ACCESS_2024_3523637 crossref_primary_10_1109_TASE_2024_3516378 crossref_primary_10_1177_02783649251360254 crossref_primary_10_1016_j_neunet_2024_106492 crossref_primary_10_1007_s42484_023_00130_3 crossref_primary_10_1109_TNNLS_2024_3453973 crossref_primary_10_1002_widm_1526 crossref_primary_10_1016_j_ecoinf_2024_102815 crossref_primary_10_1016_j_joitmc_2024_100385 crossref_primary_10_1007_s11081_024_09896_2 crossref_primary_10_1016_j_knosys_2022_110022 crossref_primary_10_1038_s42005_023_01390_y crossref_primary_10_1145_3714478 crossref_primary_10_1109_ACCESS_2021_3050841 crossref_primary_10_1007_s11063_024_11709_7 crossref_primary_10_1016_j_eswa_2025_129278 crossref_primary_10_1109_TITS_2024_3379993 crossref_primary_10_1109_TPAMI_2021_3057446 crossref_primary_10_1109_TCDS_2023_3284070 crossref_primary_10_1007_s12369_025_01214_9 crossref_primary_10_1016_j_inffus_2024_102640 crossref_primary_10_1109_TNNLS_2023_3308900 crossref_primary_10_1109_TITS_2020_3024216 crossref_primary_10_1109_TPAMI_2023_3262853 crossref_primary_10_1016_j_inffus_2022_09_007 crossref_primary_10_1109_ACCESS_2024_3402089 crossref_primary_10_1109_TASE_2025_3582733 crossref_primary_10_1109_TSMC_2020_3031086 crossref_primary_10_7717_peerj_cs_633 crossref_primary_10_1016_j_asoc_2024_111750 crossref_primary_10_1109_TCDS_2024_3485482 crossref_primary_10_3389_fpls_2023_1283055 crossref_primary_10_1016_j_jii_2024_100703 crossref_primary_10_20965_jrm_2024_p0973 crossref_primary_10_3389_frobt_2025_1605652 crossref_primary_10_3390_s23031611 crossref_primary_10_1109_LRA_2021_3056071 crossref_primary_10_1145_3533432 crossref_primary_10_1007_s10462_022_10294_2 crossref_primary_10_1016_j_knosys_2024_112575 crossref_primary_10_1007_s43154_023_00103_1 crossref_primary_10_1016_j_inffus_2023_101805 crossref_primary_10_1016_j_iot_2023_100886 crossref_primary_10_1016_j_neunet_2022_08_002 crossref_primary_10_1016_j_neunet_2023_03_025 crossref_primary_10_1007_s10994_023_06330_z crossref_primary_10_1016_j_inffus_2024_102730 crossref_primary_10_1109_LRA_2023_3291924 crossref_primary_10_1016_j_knosys_2025_113704 crossref_primary_10_1088_1361_6501_ad9347 crossref_primary_10_1016_j_ins_2021_04_062 crossref_primary_10_1109_LRA_2022_3148475 crossref_primary_10_1007_s11370_025_00629_7 crossref_primary_10_1016_j_cie_2022_108581 crossref_primary_10_1016_j_patrec_2021_08_026 crossref_primary_10_1145_3689626 crossref_primary_10_1016_j_imavis_2022_104426 crossref_primary_10_1109_TNNLS_2024_3476068 crossref_primary_10_1016_j_neucom_2024_128217 crossref_primary_10_1016_j_neunet_2024_106920 crossref_primary_10_7746_jkros_2024_19_4_369 crossref_primary_10_1007_s00521_022_07600_y crossref_primary_10_1016_j_neunet_2023_05_006 crossref_primary_10_1177_09596518231153445 crossref_primary_10_1007_s10845_021_01793_0 crossref_primary_10_1016_j_artint_2021_103635 crossref_primary_10_1007_s11370_023_00481_7 crossref_primary_10_1109_ACCESS_2025_3596836 crossref_primary_10_1109_ACCESS_2025_3573041 crossref_primary_10_1109_LRA_2024_3487484 crossref_primary_10_1007_s10846_024_02067_6 crossref_primary_10_1016_j_genrep_2025_102314 crossref_primary_10_1016_j_patcog_2025_112352 crossref_primary_10_1007_s10846_022_01603_6 crossref_primary_10_1145_3705725 crossref_primary_10_1109_TII_2024_3435499 crossref_primary_10_1109_TC_2025_3543682 crossref_primary_10_1109_MIE_2020_3034884 crossref_primary_10_1016_j_inffus_2023_102021 crossref_primary_10_1145_3657030 crossref_primary_10_3390_s23198229 crossref_primary_10_1109_ACCESS_2022_3155451 crossref_primary_10_3390_machines12020105 crossref_primary_10_1016_j_neucom_2024_127935 crossref_primary_10_3390_agriculture15091006 crossref_primary_10_3389_fnbot_2021_669534 crossref_primary_10_1007_s10015_025_01003_7 crossref_primary_10_1145_3716629 crossref_primary_10_1002_aisy_202400344 crossref_primary_10_1109_ACCESS_2023_3303869 crossref_primary_10_3390_e22111190 crossref_primary_10_1038_s41598_023_37032_0 crossref_primary_10_1371_journal_pone_0267301 crossref_primary_10_1109_ACCESS_2025_3555750 crossref_primary_10_1038_s41928_023_01054_3 crossref_primary_10_1109_TNNLS_2022_3162241 crossref_primary_10_1002_rnc_7380 crossref_primary_10_1109_TCSVT_2022_3206865 crossref_primary_10_1016_j_aap_2024_107645 crossref_primary_10_1109_TNSM_2025_3538895 crossref_primary_10_30983_icmie_v2i1_81 crossref_primary_10_1109_TCDS_2022_3200839 crossref_primary_10_1109_MCI_2022_3222049 crossref_primary_10_1016_j_adhoc_2023_103180 crossref_primary_10_1109_JIOT_2024_3407584 crossref_primary_10_1007_s10462_023_10609_x crossref_primary_10_1007_s11042_021_11219_x crossref_primary_10_1016_j_tins_2022_06_002 crossref_primary_10_1109_TPAMI_2025_3529038 crossref_primary_10_1007_s13198_025_02755_y crossref_primary_10_1016_j_neunet_2023_04_010 crossref_primary_10_1109_TCDS_2023_3276597 crossref_primary_10_1016_j_neucom_2024_127960 crossref_primary_10_1109_ACCESS_2024_3431885 crossref_primary_10_1109_TNNLS_2024_3440498 |
| Cites_doi | 10.1111/cdep.12282 10.1023/A:1022628806385 10.1109/ICRA.2018.8460934 10.1038/nature14236 10.1109/IJCNN.2017.7965898 10.1177/0278364918770733 10.3389/fnbot.2018.00059 10.1613/jair.3912 10.1016/S0079-7421(08)60536-8 10.1073/pnas.1611835114 10.1016/j.neucom.2008.10.022 10.1126/science.aab3050 10.1007/978-3-319-58347-1_1 10.1145/3293318 10.1371/journal.pone.0187736 10.1109/JRPROC.1961.287775 10.1080/09540090310001655110 10.1016/j.neunet.2019.01.012 10.1109/TEVC.2006.890271 10.1177/0278364914546030 10.1016/S1364-6613(99)01294-2 10.1371/journal.pcbi.1006713 10.1080/09540099550039318 10.1016/j.cviu.2018.10.009 10.1002/acp.1533 10.1109/TAMD.2010.2056368 10.1037/0033-295X.102.3.419 10.1109/MSP.2014.2347059 10.1162/1064546053278973 10.3389/fnbot.2018.00078 10.1109/TPAMI.2006.79 10.1007/s10514-015-9459-7 10.1016/j.neunet.2018.07.006 10.1126/science.291.5504.599 10.1080/08982119808919168 10.1007/s12559-016-9389-5 10.1109/IJCNN.2010.5596771 10.3389/frobt.2018.00012 10.1016/j.neunet.2019.03.010 10.1016/j.robot.2018.11.012 10.1007/s11263-014-0733-5 |
| ContentType | Journal Article |
| Copyright | 2019 Attribution - NonCommercial |
| Copyright_xml | – notice: 2019 – notice: Attribution - NonCommercial |
| DBID | AAYXX CITATION 1XC VOOES |
| DOI | 10.1016/j.inffus.2019.12.004 |
| DatabaseName | CrossRef Hyper Article en Ligne (HAL) Hyper Article en Ligne (HAL) (Open Access) |
| DatabaseTitle | CrossRef |
| DatabaseTitleList | |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Mathematics Computer Science |
| EISSN | 1872-6305 |
| EndPage | 68 |
| ExternalDocumentID | oai:HAL:hal-02381343v1 10_1016_j_inffus_2019_12_004 S1566253519307377 |
| GroupedDBID | --K --M .DC .~1 0R~ 1B1 1~. 1~5 29I 4.4 457 4G. 5GY 5VS 7-5 71M 8P~ AACTN AAEDT AAEDW AAIAV AAIKJ AAKOC AALRI AAOAW AAQFI AAQXK AAXUO AAYFN ABBOA ABFNM ABJNI ABMAC ABXDB ABYKQ ACDAQ ACGFS ACNNM ACRLP ACZNC ADBBV ADEZE ADJOM ADMUD ADTZH AEBSH AECPX AEKER AENEX AFKWA AFTJW AGHFR AGUBO AGYEJ AHJVU AHZHX AIALX AIEXJ AIKHN AITUG AJBFU AJOXV ALMA_UNASSIGNED_HOLDINGS AMFUW AMRAJ AOUOD ASPBG AVWKF AXJTR AZFZN BJAXD BKOJK BLXMC CS3 DU5 EBS EFJIC EFLBG EJD EO8 EO9 EP2 EP3 F0J F5P FDB FEDTE FGOYB FIRID FNPLU FYGXN G-Q GBLVA GBOLZ HVGLF HZ~ IHE J1W JJJVA KOM M41 MO0 N9A O-L O9- OAUVE OZT P-8 P-9 P2P PC. Q38 R2- RIG ROL RPZ SDF SDG SDP SES SEW SPC SPCBC SST SSV SSZ T5K UHS ZMT ~G- 9DU AATTM AAXKI AAYWO AAYXX ABWVN ACLOT ACRPL ACVFH ADCNI ADNMO AEIPS AEUPX AFJKZ AFPUW AGQPQ AIGII AIIUN AKBMS AKRWK AKYEP ANKPU APXCP CITATION EFKBS ~HD 1XC VOOES |
| ID | FETCH-LOGICAL-c452t-8b9c10beff882ebc8594cc4cde144584af1e7ca746a2ee1c6c0449f2697ae18b3 |
| ISICitedReferencesCount | 339 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000516799200005&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1566-2535 |
| IngestDate | Tue Oct 14 20:45:54 EDT 2025 Sat Nov 29 07:05:11 EST 2025 Tue Nov 18 22:16:45 EST 2025 Fri Feb 23 02:47:02 EST 2024 |
| IsDoiOpenAccess | true |
| IsOpenAccess | true |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | Reinforcement Learning Continual Learning Catastrophic Forgetting Lifelong Learning Robotics Deep Learning ContinualLearning |
| Language | English |
| License | Attribution - NonCommercial: http://creativecommons.org/licenses/by-nc |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-c452t-8b9c10beff882ebc8594cc4cde144584af1e7ca746a2ee1c6c0449f2697ae18b3 |
| ORCID | 0000-0002-5739-1618 0000-0003-3362-9326 |
| OpenAccessLink | https://hal.science/hal-02381343 |
| PageCount | 17 |
| ParticipantIDs | hal_primary_oai_HAL_hal_02381343v1 crossref_citationtrail_10_1016_j_inffus_2019_12_004 crossref_primary_10_1016_j_inffus_2019_12_004 elsevier_sciencedirect_doi_10_1016_j_inffus_2019_12_004 |
| PublicationCentury | 2000 |
| PublicationDate | 2020-06-01 |
| PublicationDateYYYYMMDD | 2020-06-01 |
| PublicationDate_xml | – month: 06 year: 2020 text: 2020-06-01 day: 01 |
| PublicationDecade | 2020 |
| PublicationTitle | Information fusion |
| PublicationYear | 2020 |
| Publisher | Elsevier B.V Elsevier |
| Publisher_xml | – name: Elsevier B.V – name: Elsevier |
| References | Mermillod, Bugaiska, Bonin (bib0103) 2013; 4 LeCun, Cortes (bib0079) 2010 Maltoni, Lomonaco (bib0097) 2019; 116 Aljundi, Chakravarty, Tuytelaars (bib0003) 2017 Colas, Sigaud, Oudeyer (bib0027) 2018; 80 Oudeyer (bib0112) 2018; abs/1802.10546 Borji (bib0011) 2018; 179 Lyubova, Ivaldi, Filliat (bib0094) 2015 Romeres, Zorzi, Camoriano, Chiuso (bib0133) 2016 Draelos, Miner, Lamb, Cox, Vineyard, Carlson, Severa, James, Aimone (bib0037) 2017 Netzer, Wang, Coates, Bissacco, Wu, Ng (bib0110) 2011 Bühlmann, van de Geer (bib0014) 2011 Yoon, Yang, Lee, Hwang (bib0174) 2018 Mankowitz, Žídek, Barreto, Horgan, Hessel, Quan, Oh, van Hasselt, Silver, Schaul (bib0099) 2018 Moens, Zénon (bib0109) 2019; 15 Pasquale, Ciliberto, Rosasco, Natale (bib0117) 2016 Traoré, Caselles-Dupré, Lesort, Sun, Cai, Rodríguez, Filliat (bib0160) 2019; abs/1907.05855 Collet, Xiong, Gurau, Hebert, Srinivasa (bib0028) 2015; 34 Pasquale, Ciliberto, Odone, Rosasco, Natale (bib0116) 2015 Michieli, Zanuttigh (bib0104) 2019 Carlson, Betteridge, Kisiel, Settles, Hruschka Jr, Mitchell (bib0019) 2010; 5 Lesort, Caselles-Dupré, Garcia-Ortiz, Goudou, Filliat (bib0081) 2019 Brazdil, Giraud-Carrier, Soares, Vilalta (bib0012) 2008 Craye, Lesort, Filliat, Goudou (bib0030) 2019; 112 Levine, Finn, Darrell, Abbeel (bib0086) 2016; 17 Zhao, Hospedales, Stulp, Sigaud (bib0177) 2017 Gepperth, Karaoguz (bib0051) 2016; 8 Krizhevsky, Hinton (bib0074) 2009 Ramapuram, Gregorova, Kalousis (bib0124) 2017 Russell, Norvig (bib0135) 2009 Robins (bib0132) 1995; 7 Chaudhry, Dokania, Ajanthan, Torr (bib0022) 2018 Mattner, Lange, Riedmiller (bib0100) 2012 Käding, Rodner, Freytag, Denzler (bib0066) 2016 Kim, Coninx, Doncieux (bib0072) 2019; abs/1901.00811 McCloskey, Cohen (bib0102) 1989; 24 Im, Ma, Taylor, Branson (bib0061) 2018 Bengio, Louradour, Collobert, Weston (bib0009) 2009 Kemker, McClure, Abitino, Hayes, Kanan (bib0071) 2017 Shelhamer, Mahmoudieh, Argus, Darrell (bib0149) 2016 . Parisi, Tani, Weber, Wermter (bib0115) 2018; 12 Pentina, Lampert (bib0119) 2015 Chaudhry, Ranzato, Rohrbach, Elhoseiny (bib0023) 2019 Kemker, Kanan (bib0070) 2018 Lake, Salakhutdinov, Tenenbaum (bib0077) 2015; 350 Delvenne (bib0032) 2009; 23 Finn, Tan, Duan, Darrell, Levine, Abbeel (bib0045) 2015 Yu, Zhang, Song, Seff, Xiao (bib0175) 2015; abs/1506.03365 Hayes, Kemker, Cahill, Kanan (bib0056) 2018 Li, Hoiem (bib0087) 2017 Oudeyer, Kaplan, Hafner (bib0113) 2007; 11 Pratt (bib0122) 1993 Bojarski, Testa, Dworakowski, Firner, Flepp, Goyal, Jackel, Monfort, Muller, Zhang, Zhang, Zhao, Zieba (bib0010) 2016; abs/1604.07316 Kirkpatrick, Pascanu, Rabinowitz, Veness, Desjardins, Rusu, Milan, Quan, Ramalho, Grabska-Barwinska (bib0073) 2017 Ring (bib0128) 1994 Rusu, Gomez Colmenarejo, Gulcehre, Desjardins, Kirkpatrick, Pascanu, Mnih, Kavukcuoglu, Hadsell (bib0136) 2015 Wong (bib0170) 2016 Gandhi, Pinto, Gupta (bib0049) 2017 Zenke, Poole, Ganguli (bib0176) 2017; 70 Doncieux, Filliat, Díaz-Rodríguez, Hospedales, Duro, Coninx, Roijers, Girard, Perrin, Sigaud (bib0036) 2018 French (bib0047) 1999; 3 Serra, Suris, Miron, Karatzoglou (bib0146) 2018; 80 Achille, Eccles, Matthey, Burgess, Watters, Lerchner, Higgins (bib0001) 2018 Kamra, Gupta, Liu (bib0069) 2017 Everingham, Eslami, Van Gool, Williams, Winn, Zisserman (bib0039) 2015; 111 McClelland, McNaughton, O’reilly (bib0101) 1995; 102 Agrawal, Nair, Abbeel, Malik, Levine (bib0002) 2016 Sutton, Barto (bib0157) 1998; 2 Furlanello, Zhao, Saxe, Itti, Tjan (bib0048) 2016 Lake, Salakhutdinov, Gross, Tenenbaum (bib0076) 2011; 33 Romeres, Zorzi, Camoriano, Traversaro, Chiuso (bib0134) 2019 Seff, Beatson, Suo, Liu (bib0145) 2017; abs/1705.08395 Dhar, Singh, Peng, Wu, Chellappa (bib0033) 2019 Sodhani, Chandar, Bengio (bib0153) 2018; abs/1811.07017 Krizhevsky, Sutskever, Hinton (bib0075) 2012 Parisi, Kemker, Part, Kanan, Wermter (bib0114) 2019; 113 Mallya, Lazebnik (bib0096) 2018 Rusu, Rabinowitz, Desjardins, Soyer, Kirkpatrick, Kavukcuoglu, Pascanu, Hadsell (bib0137) 2016 Pfulb, Gepperth (bib0120) 2019 Codevilla, Muller, Dosovitskiy, L’opez, Koltun (bib0025) 2017 Gepperth, Hammer (bib0050) 2016 Pinto, Gupta (bib0121) 2015 Fei-Fei, Fergus, Perona (bib0042) 2006; 28 Welinder, Branson, Mita, Wah, Schroff, Belongie, Perona (bib0168) 2010 Schmidhuber (bib0142) 2010; 2 Settles (bib0148) 2009 Lomonaco, Maltoni (bib0090) 2016 Minsky (bib0106) 1961; 49 Jonschkowski, Brock (bib0064) 2015; 39 Lesort, Díaz-Rodríguez, Goudou, Filliat (bib0082) 2018 Salimans, Goodfellow, Zaremba, Cheung, Radford, Chen, Chen (bib0139) 2016 Lesort, Gepperth, Stoian, Filliat (bib0083) 2019 Jaritz, de Charette, Toromanoff, Perot, Nashashibi (bib0063) 2018 Chen, Liu (bib0024) 2018; 12 Hinton, Vinyals, Dean (bib0060) 2015 Sarkar, Meeker (bib0140) 1998; 10 Turing (bib0162) 2009 Cavallari, Golodetz, Lord, Valentin, Di Stefano, Torr (bib0021) 2017 Finn, Abbeel, Levine (bib0044) 2017 Kalifou, Caselles-Dupré, Lesort, Sun, Diaz-Rodriguez, Filliat (bib0068) 2019 Rhinehart, McAllister, Levine (bib0126) 2018; abs/1810.06544 Mitchell, Cohen, Hruscha, Talukdar, Betteridge, Carlson, Dalvi, Gardner, Kisiel, Krishnamurthy, Lao, Mazaitis, Mohammad, Nakashole, Platanios, Ritter, Samadi, Settles, Wang, Wijaya, Gupta, Chen, Saparov, Greaves, Welling (bib0107) 2015 Xiao, Rasul, Vollgraf (bib0173) 2017 Sun, Gomez, Schmidhuber (bib0156) 2011 Craye, Filliat, Goudou (bib0029) 2015 G. Csurka, A Comprehensive Survey on Domain Adaptation for Visual Applications, Springer International Publishing, Cham, pp. 1–35. doi Díaz-Rodríguez, Lomonaco, Filliat, Maltoni (bib0034) 2018 Weng, McClelland, Pentland, Sporns, Stockman, Sur, Thelen (bib0169) 2001; 291 Zhou, Sohn, Lee (bib0178) 2012; 22 Wang, Ramanan, Hebert (bib0167) 2017 Burda, Edwards, Pathak, Storkey, Darrell, Efros (bib0015) 2019 Hayes, Cahill, Kanan (bib0055) 2018 Lee, Kim, Jun, Ha, Zhang (bib0080) 2017 Goodfellow, Pouget-Abadie, Mirza, Xu, Warde-Farley, Ozair, Courville, Bengio (bib0052) 2014 He, Shen, Cui (bib0058) 2019 Heusel, Ramsauer, Unterthiner, Nessler, Hochreiter (bib0059) 2017 Lesort, Stoian, Goudou, Filliat (bib0085) 2019 Jung, Ju, Jung, Kim (bib0065) 2016; abs/1607.00122 Bellas, Becerra, Duro (bib0005) 2009; 72 Kalashnikov, Irpan, Pastor, Ibarz, Herzog, Jang, Quillen, Holly, Kalakrishnan, Vanhoucke, Levine (bib0067) 2018; 87 Rios, Itti (bib0130) 2019 Rusu, Vecerik, Rothörl, Heess, Pascanu, Hadsell (bib0138) 2016; abs/1610.04286 Settles (bib0147) 2009 Valiant (bib0163) 1984 Duan (bib0038) 2017 U. Michieli, P. Zanuttigh, Knowledge distillation for incremental learning in semantic segmentation, Provided by the SAO/NASA Astrophysics Data System. Triki Rannen, Aljundi, Blaschko, Tuytelaars (bib0161) 2017 Lungarella, Metta, Pfeifer, Sandini (bib0093) 2003; 15 Velez, Clune (bib0165) 2017; 12 Lesort, Seurin, Li, Díaz-Rodríguez, Filliat (bib0084) 2019 Thrun, Mitchell (bib0159) 1995 Farquhar, Gal (bib0041) 2018 Goodfellow, Mirza, Xiao, Courville, Bengio (bib0053) 2013 He, Jaeger (bib0057) 2018 Shmelkov, Schmid, Alahari (bib0151) 2017 Schwarz, Czarnecki, Luketina, Grabska-Barwinska, Teh, Pascanu, Hadsell (bib0144) 2018; 80 Schaul, Quan, Antonoglou, Silver (bib0141) 2015 Wu, Herranz, Liu, wang, van de Weijer, Raducanu (bib0171) 2018 Lopez-Paz, Ranzato (bib0092) 2017 Camoriano, Traversaro, Rosasco, Metta, Nori (bib0017) 2016 Caselles-Dupré, Garcia-Ortiz, Filliat (bib0020) 2019 Colas, Fournier, Chetouani, Sigaud, Oudeyer (bib0026) 2019; 97 Fan, Zhu, Zhu, Liu, Zeng, Gupta, Creus-Costa, Savarese, Fei-Fei (bib0040) 2018 Wang, Zheng, Yu, Miao (bib0166) 2019; 10 Mallya, Davis, Lazebnik (bib0095) 2018 Azagra, Golemo, Mollard, Lopes, Civera, Murillo (bib0004) 2017 Nguyen, Li, Bui, Turner (bib0111) 2018 Stulp, Herlant, Hoarau, Raiola (bib0155) 2014 Sünderhauf, Brock, Scheirer, Hadsell, Fox, Leitner, Upcroft, Abbeel, Burgard, Milford, Corke (bib0158) 2018; 37 Schulman, Wolski, Dhariwal, Radford, Klimov (bib0143) 2017 Patel, Gopalan, Li, Chellappa (bib0118) 2015; 32 F. Bellas, A. Faiña, G. Varela, R.J. Duro, A cognitive developmental robotics architecture for lifelong learning by evolution in real robots. The 2010 International Joint Conference on Neural Networks (IJCNN). IEEE, 2010. Lomonaco, Maltoni (bib0091) 2017; 78 Ritter, Botev, Barber (bib0131) 2018 Bredeche, Haasdijk, Prieto (bib0013) 2018; 5 van Hoof, Chen, Karl, van der Smagt, Peters (bib0164) 2016 2019. Camoriano, Pasquale, Ciliberto, Natale, Rosasco, Metta (bib0016) 2017 Doersch, Zisserman (bib0035) 2017 Riemer, Cases, Ajemian, Liu, Rish, Tu, Tesauro (bib0127) 2019 Rebuffi, Kolesnikov, Sperl, Lampert (bib0125) 2017 Mandlekar, Zhu, Garg, Booher, Spero, Tung, Gao, Emmons, Gupta, Orbay, Savarese, Fei-Fei (bib0098) 2018 Raffin, Hill, Traoré, Lesort, Díaz-Rodríguez, Filliat (bib0123) 2019 Cangelosi, Schlesinger (bib0018) 2018 Mnih, Kavukcuoglu, Silver, Rusu, Veness, Bellemare, Graves, Riedmiller, Fidjeland, Ostrovski, Petersen, Beattie, Sadik, Antonoglou, King, Kumaran, Wierstra, Legg, Hassabis (bib0108) 2015; 518 Wu, Chen, Wang, Ye, Liu, Guo, Zhang, Fu (bib0172) 2018; abs/1802.00853 Bellas, Duro, Faina, Souto (bib0006) 2010; 2 Ring (bib0129) 2005 Sprechmann, Jayakumar, Rae, Pritzel, Badia, Uria, Vinyals, Hassabis, Pascanu, Blundell (bib0154) 2018 Gopnik, Meltzoff, Kuhl (bib0054) 2001; 189 Smith, Gasser (bib0152) 2005; 11 Fernando, Banarse, Blundell, Zwols, Ha, Rusu, Pritzel, Wierstra (bib0043) 2017; abs/1701.08734 Laversanne-Finot, Pere, Oudeyer (bib0078) 2018; 87 Lomonaco, Desai, Culurciello, Maltoni (bib0089) 2019 Jaeger (bib0062) 2017; 18 Shin, Lee, Kim, Kim (bib0150) 2017 Forestier, Mollard, Oudeyer (bib0046) 2017 Lin (bib0088) 1992; 8 Bellemare, Naddaf, Veness, Bowling (bib0008) 2013; 47 Shelhamer (10.1016/j.inffus.2019.12.004_bib0149) 2016 Settles (10.1016/j.inffus.2019.12.004_bib0147) 2009 Ritter (10.1016/j.inffus.2019.12.004_bib0131) 2018 Collet (10.1016/j.inffus.2019.12.004_bib0028) 2015; 34 Fei-Fei (10.1016/j.inffus.2019.12.004_bib0042) 2006; 28 Gandhi (10.1016/j.inffus.2019.12.004_bib0049) 2017 Camoriano (10.1016/j.inffus.2019.12.004_bib0016) 2017 Hayes (10.1016/j.inffus.2019.12.004_bib0055) 2018 Zhao (10.1016/j.inffus.2019.12.004_bib0177) 2017 Pasquale (10.1016/j.inffus.2019.12.004_bib0116) 2015 Bellas (10.1016/j.inffus.2019.12.004_bib0006) 2010; 2 10.1016/j.inffus.2019.12.004_bib0031 Lomonaco (10.1016/j.inffus.2019.12.004_bib0091) 2017; 78 Rhinehart (10.1016/j.inffus.2019.12.004_bib0126) 2018; abs/1810.06544 Bellas (10.1016/j.inffus.2019.12.004_bib0005) 2009; 72 Achille (10.1016/j.inffus.2019.12.004_bib0001) 2018 Xiao (10.1016/j.inffus.2019.12.004_bib0173) 2017 Laversanne-Finot (10.1016/j.inffus.2019.12.004_bib0078) 2018; 87 Agrawal (10.1016/j.inffus.2019.12.004_bib0002) 2016 Borji (10.1016/j.inffus.2019.12.004_bib0011) 2018; 179 Seff (10.1016/j.inffus.2019.12.004_bib0145) 2017; abs/1705.08395 Kirkpatrick (10.1016/j.inffus.2019.12.004_bib0073) 2017 McClelland (10.1016/j.inffus.2019.12.004_bib0101) 1995; 102 Valiant (10.1016/j.inffus.2019.12.004_bib0163) 1984 Bellemare (10.1016/j.inffus.2019.12.004_bib0008) 2013; 47 Jaritz (10.1016/j.inffus.2019.12.004_bib0063) 2018 Sutton (10.1016/j.inffus.2019.12.004_bib0157) 1998; 2 Finn (10.1016/j.inffus.2019.12.004_bib0045) 2015 Krizhevsky (10.1016/j.inffus.2019.12.004_bib0075) 2012 Pinto (10.1016/j.inffus.2019.12.004_bib0121) 2015 Pfulb (10.1016/j.inffus.2019.12.004_bib0120) 2019 Chaudhry (10.1016/j.inffus.2019.12.004_bib0023) 2019 Craye (10.1016/j.inffus.2019.12.004_bib0030) 2019; 112 Duan (10.1016/j.inffus.2019.12.004_bib0038) 2017 Gepperth (10.1016/j.inffus.2019.12.004_sbref0048) 2016 Lopez-Paz (10.1016/j.inffus.2019.12.004_bib0092) 2017 Nguyen (10.1016/j.inffus.2019.12.004_bib0111) 2018 Jonschkowski (10.1016/j.inffus.2019.12.004_bib0064) 2015; 39 Thrun (10.1016/j.inffus.2019.12.004_bib0159) 1995 Rios (10.1016/j.inffus.2019.12.004_bib0130) 2019 Mattner (10.1016/j.inffus.2019.12.004_bib0100) 2012 Rusu (10.1016/j.inffus.2019.12.004_sbref0133) 2015 Caselles-Dupré (10.1016/j.inffus.2019.12.004_sbref0019) 2019 Maltoni (10.1016/j.inffus.2019.12.004_bib0097) 2019; 116 Hayes (10.1016/j.inffus.2019.12.004_bib0056) 2018 Kamra (10.1016/j.inffus.2019.12.004_bib0069) 2017 Mallya (10.1016/j.inffus.2019.12.004_bib0096) 2018 Im (10.1016/j.inffus.2019.12.004_bib0061) 2018 Jung (10.1016/j.inffus.2019.12.004_bib0065) 2016; abs/1607.00122 LeCun (10.1016/j.inffus.2019.12.004_bib0079) 2010 Lesort (10.1016/j.inffus.2019.12.004_bib0082) 2018 Mitchell (10.1016/j.inffus.2019.12.004_sbref0104) 2015 Levine (10.1016/j.inffus.2019.12.004_bib0086) 2016; 17 Fan (10.1016/j.inffus.2019.12.004_bib0040) 2018 Weng (10.1016/j.inffus.2019.12.004_bib0169) 2001; 291 Salimans (10.1016/j.inffus.2019.12.004_bib0139) 2016 van Hoof (10.1016/j.inffus.2019.12.004_bib0164) 2016 Draelos (10.1016/j.inffus.2019.12.004_bib0037) 2017 Furlanello (10.1016/j.inffus.2019.12.004_bib0048) 2016 Zenke (10.1016/j.inffus.2019.12.004_bib0176) 2017; 70 He (10.1016/j.inffus.2019.12.004_bib0057) 2018 Schulman (10.1016/j.inffus.2019.12.004_bib0143) 2017 Kemker (10.1016/j.inffus.2019.12.004_bib0071) 2017 Wang (10.1016/j.inffus.2019.12.004_bib0166) 2019; 10 Lesort (10.1016/j.inffus.2019.12.004_bib0083) 2019 Azagra (10.1016/j.inffus.2019.12.004_sbref0004) 2017 Gepperth (10.1016/j.inffus.2019.12.004_bib0051) 2016; 8 Ramapuram (10.1016/j.inffus.2019.12.004_bib0124) 2017 Moens (10.1016/j.inffus.2019.12.004_bib0109) 2019; 15 Jaeger (10.1016/j.inffus.2019.12.004_bib0062) 2017; 18 Schaul (10.1016/j.inffus.2019.12.004_bib0141) 2015 Welinder (10.1016/j.inffus.2019.12.004_bib0168) 2010 Doncieux (10.1016/j.inffus.2019.12.004_bib0036) 2018 Russell (10.1016/j.inffus.2019.12.004_bib0135) 2009 Dhar (10.1016/j.inffus.2019.12.004_bib0033) 2019 Ring (10.1016/j.inffus.2019.12.004_bib0129) 2005 Lyubova (10.1016/j.inffus.2019.12.004_bib0094) 2015 Serra (10.1016/j.inffus.2019.12.004_bib0146) 2018; 80 Sun (10.1016/j.inffus.2019.12.004_bib0156) 2011 McCloskey (10.1016/j.inffus.2019.12.004_bib0102) 1989; 24 Smith (10.1016/j.inffus.2019.12.004_bib0152) 2005; 11 Stulp (10.1016/j.inffus.2019.12.004_bib0155) 2014 Oudeyer (10.1016/j.inffus.2019.12.004_bib0113) 2007; 11 Schwarz (10.1016/j.inffus.2019.12.004_bib0144) 2018; 80 Lake (10.1016/j.inffus.2019.12.004_bib0076) 2011; 33 Mallya (10.1016/j.inffus.2019.12.004_bib0095) 2018 Shmelkov (10.1016/j.inffus.2019.12.004_bib0151) 2017 Carlson (10.1016/j.inffus.2019.12.004_bib0019) 2010; 5 Krizhevsky (10.1016/j.inffus.2019.12.004_bib0074) 2009 Parisi (10.1016/j.inffus.2019.12.004_bib0115) 2018; 12 10.1016/j.inffus.2019.12.004_bib0105 Yoon (10.1016/j.inffus.2019.12.004_bib0174) 2018 Lesort (10.1016/j.inffus.2019.12.004_bib0084) 2019 Robins (10.1016/j.inffus.2019.12.004_bib0132) 1995; 7 Aljundi (10.1016/j.inffus.2019.12.004_bib0003) 2017 French (10.1016/j.inffus.2019.12.004_bib0047) 1999; 3 He (10.1016/j.inffus.2019.12.004_sbref0056) 2019 Triki Rannen (10.1016/j.inffus.2019.12.004_bib0161) 2017 Cavallari (10.1016/j.inffus.2019.12.004_bib0021) 2017 Burda (10.1016/j.inffus.2019.12.004_bib0015) 2019 Patel (10.1016/j.inffus.2019.12.004_bib0118) 2015; 32 Doersch (10.1016/j.inffus.2019.12.004_bib0035) 2017 Finn (10.1016/j.inffus.2019.12.004_bib0044) 2017 Rusu (10.1016/j.inffus.2019.12.004_bib0137) 2016 Sünderhauf (10.1016/j.inffus.2019.12.004_bib0158) 2018; 37 Wu (10.1016/j.inffus.2019.12.004_bib0171) 2018 Sprechmann (10.1016/j.inffus.2019.12.004_bib0154) 2018 Riemer (10.1016/j.inffus.2019.12.004_bib0127) 2019 Mankowitz (10.1016/j.inffus.2019.12.004_bib0099) 2018 Colas (10.1016/j.inffus.2019.12.004_bib0026) 2019; 97 Forestier (10.1016/j.inffus.2019.12.004_bib0046) 2017 Schmidhuber (10.1016/j.inffus.2019.12.004_bib0142) 2010; 2 Rebuffi (10.1016/j.inffus.2019.12.004_bib0125) 2017 Velez (10.1016/j.inffus.2019.12.004_bib0165) 2017; 12 Michieli (10.1016/j.inffus.2019.12.004_bib0104) 2019 Lomonaco (10.1016/j.inffus.2019.12.004_bib0089) 2019 Sarkar (10.1016/j.inffus.2019.12.004_bib0140) 1998; 10 Yu (10.1016/j.inffus.2019.12.004_bib0175) 2015; abs/1506.03365 Chen (10.1016/j.inffus.2019.12.004_bib0024) 2018; 12 Wong (10.1016/j.inffus.2019.12.004_bib0170) 2016 Craye (10.1016/j.inffus.2019.12.004_bib0029) 2015 Bredeche (10.1016/j.inffus.2019.12.004_bib0013) 2018; 5 Kemker (10.1016/j.inffus.2019.12.004_bib0070) 2018 Goodfellow (10.1016/j.inffus.2019.12.004_bib0052) 2014 Lungarella (10.1016/j.inffus.2019.12.004_bib0093) 2003; 15 Li (10.1016/j.inffus.2019.12.004_bib0087) 2017 Codevilla (10.1016/j.inffus.2019.12.004_bib0025) 2017 10.1016/j.inffus.2019.12.004_bib0007 Bengio (10.1016/j.inffus.2019.12.004_bib0009) 2009 Wang (10.1016/j.inffus.2019.12.004_bib0167) 2017 Chaudhry (10.1016/j.inffus.2019.12.004_bib0022) 2018 Kalifou (10.1016/j.inffus.2019.12.004_bib0068) 2019 Mermillod (10.1016/j.inffus.2019.12.004_bib0103) 2013; 4 Parisi (10.1016/j.inffus.2019.12.004_bib0114) 2019; 113 Traoré (10.1016/j.inffus.2019.12.004_bib0160) 2019; abs/1907.05855 Romeres (10.1016/j.inffus.2019.12.004_bib0133) 2016 Mandlekar (10.1016/j.inffus.2019.12.004_bib0098) 2018 Kalashnikov (10.1016/j.inffus.2019.12.004_bib0067) 2018; 87 Shin (10.1016/j.inffus.2019.12.004_bib0150) 2017 Camoriano (10.1016/j.inffus.2019.12.004_bib0017) 2016 Turing (10.1016/j.inffus.2019.12.004_bib0162) 2009 Lesort (10.1016/j.inffus.2019.12.004_sbref0079) 2019 Oudeyer (10.1016/j.inffus.2019.12.004_bib0112) 2018; abs/1802.10546 Sodhani (10.1016/j.inffus.2019.12.004_bib0153) 2018; abs/1811.07017 Delvenne (10.1016/j.inffus.2019.12.004_bib0032) 2009; 23 Zhou (10.1016/j.inffus.2019.12.004_bib0178) 2012; 22 Bojarski (10.1016/j.inffus.2019.12.004_bib0010) 2016; abs/1604.07316 Lesort (10.1016/j.inffus.2019.12.004_bib0085) 2019 Wu (10.1016/j.inffus.2019.12.004_bib0172) 2018; abs/1802.00853 Cangelosi (10.1016/j.inffus.2019.12.004_bib0018) 2018 Minsky (10.1016/j.inffus.2019.12.004_bib0106) 1961; 49 Pratt (10.1016/j.inffus.2019.12.004_bib0122) 1993 Lake (10.1016/j.inffus.2019.12.004_bib0077) 2015; 350 Pentina (10.1016/j.inffus.2019.12.004_bib0119) 2015 Ring (10.1016/j.inffus.2019.12.004_bib0128) 1994 Heusel (10.1016/j.inffus.2019.12.004_bib0059) 2017 Díaz-Rodríguez (10.1016/j.inffus.2019.12.004_sbref0032) 2018 Hinton (10.1016/j.inffus.2019.12.004_bib0060) 2015 Käding (10.1016/j.inffus.2019.12.004_bib0066) 2016 Netzer (10.1016/j.inffus.2019.12.004_bib0110) 2011 Kim (10.1016/j.inffus.2019.12.004_bib0072) 2019; abs/1901.00811 Goodfellow (10.1016/j.inffus.2019.12.004_bib0053) 2013 Lee (10.1016/j.inffus.2019.12.004_bib0080) 2017 Lomonaco (10.1016/j.inffus.2019.12.004_bib0090) 2016 Fernando (10.1016/j.inffus.2019.12.004_bib0043) 2017; abs/1701.08734 Lin (10.1016/j.inffus.2019.12.004_bib0088) 1992; 8 Bühlmann (10.1016/j.inffus.2019.12.004_bib0014) 2011 Everingham (10.1016/j.inffus.2019.12.004_bib0039) 2015; 111 Settles (10.1016/j.inffus.2019.12.004_bib0148) 2009 Gopnik (10.1016/j.inffus.2019.12.004_bib0054) 2001; 189 Mnih (10.1016/j.inffus.2019.12.004_bib0108) 2015; 518 Raffin (10.1016/j.inffus.2019.12.004_bib0123) 2019 Colas (10.1016/j.inffus.2019.12.004_bib0027) 2018; 80 Pasquale (10.1016/j.inffus.2019.12.004_bib0117) 2016 Rusu (10.1016/j.inffus.2019.12.004_bib0138) 2016; abs/1610.04286 Brazdil (10.1016/j.inffus.2019.12.004_bib0012) 2008 Farquhar (10.1016/j.inffus.2019.12.004_bib0041) 2018 Romeres (10.1016/j.inffus.2019.12.004_bib0134) 2019 |
| References_xml | – year: 2010 ident: bib0079 article-title: MNIST handwritten digit database publication-title: Public – year: 2009 ident: bib0135 article-title: Artificial Intelligence: A Modern Approach – start-page: 466 year: 2019 end-page: 480 ident: bib0083 article-title: Marginal replay vs conditional replay for continual learning publication-title: Artificial Neural Networks and Machine Learning - ICANN 2019: Deep Learning - 28th International Conference on Artificial Neural Networks, Munich, Germany, September 17–19, 2019, Proceedings, Part II – start-page: 5074 year: 2016 end-page: 5082 ident: bib0002 article-title: Learning to poke by poking: experiential learning of intuitive physics publication-title: Advances in Neural Information Processing Systems 29 – volume: abs/1701.08734 year: 2017 ident: bib0043 article-title: Pathnet: evolution channels gradient descent in super neural networks publication-title: CoRR – year: 2019 ident: bib0127 article-title: Learning to learn without forgetting by maximizing transfer and minimizing interference publication-title: International Conference on Learning Representations – volume: abs/1811.07017 year: 2018 ident: bib0153 article-title: On training recurrent neural networks for lifelong learning publication-title: CoRR – volume: 17 start-page: 1334 year: 2016 end-page: 1373 ident: bib0086 article-title: End-to-end training of deep visuomotor policies publication-title: J. Mach. Learn. Res. – volume: 80 start-page: 4548 year: 2018 end-page: 4557 ident: bib0146 article-title: Overcoming catastrophic forgetting with hard attention to the task publication-title: ICML – volume: abs/1802.00853 year: 2018 ident: bib0172 article-title: Incremental classifier learning with generative adversarial networks publication-title: CoRR – volume: 2 year: 1998 ident: bib0157 article-title: Introduction to reinforcement learning – year: 2010 ident: bib0168 article-title: Caltech-UCSD Birds 200 publication-title: Technical Report – year: 2017 ident: bib0173 article-title: Fashion-mnist: a novel image dataset for benchmarking machine learning algorithms publication-title: arXiv preprint arXiv:1708.07747 – volume: 37 start-page: 405 year: 2018 end-page: 420 ident: bib0158 article-title: The limits and potentials of deep learning for robotics publication-title: Int. J. Robot. Res. – start-page: 41 year: 2009 end-page: 48 ident: bib0009 article-title: Curriculum learning publication-title: Proceedings of the 26th annual international conference on machine learning – volume: 291 start-page: 599 year: 2001 end-page: 600 ident: bib0169 article-title: Autonomous mental development by robots and animals publication-title: Science – year: 2018 ident: bib0034 article-title: Don’t forget, there is more than forgetting: new metrics for Continual Learning publication-title: Workshop on Continual Learning, NeurIPS 2018 (Neural Information Processing Systems – year: 2018 ident: bib0082 article-title: State representation learning for control: an overview publication-title: Neur. Netw. – volume: 8 start-page: 293 year: 1992 end-page: 321 ident: bib0088 article-title: Self-improving reactive agents based on reinforcement learning, planning and teaching publication-title: Mach. Learn. – volume: 116 start-page: 56 year: 2019 end-page: 73 ident: bib0097 article-title: Continuous learning in single-incremental-task scenarios publication-title: Neural Netw. – year: 2018 ident: bib0099 article-title: Unicorn: continual learning with a universal, off-policy agent publication-title: arXiv Preprint arXiv:1802.08294 – start-page: 4652 year: 2017 end-page: 4662 ident: bib0080 article-title: Overcoming catastrophic forgetting by incremental moment matching publication-title: Advances in Neural Information Processing Systems – start-page: 3406 year: 2015 end-page: 3413 ident: bib0121 article-title: Supersizing self-supervision: learning to grasp from 50k tries and 700 robot hours publication-title: 2016 IEEE Int. Conf. Robot. Automat. (ICRA) – volume: 10 start-page: 539 year: 1998 end-page: 549 ident: bib0140 article-title: A bayesian on-line change detection algorithm with process monitoring applications publication-title: Qual. Eng. – start-page: 67 year: 2018 end-page: 82 ident: bib0095 article-title: Piggyback: adapting a single network to multiple tasks by learning to mask weights publication-title: Proceedings of the European Conference on Computer Vision (ECCV) – year: 2018 ident: bib0070 article-title: Fearnet: brain-inspired model for incremental learning publication-title: International Conference on Learning Representations – year: 2009 ident: bib0074 article-title: Learning multiple layers of features from tiny images publication-title: Technical Report – year: 2019 ident: bib0104 article-title: Incremental learning techniques for semantic segmentation publication-title: International Conference on Computer Vision (ICCV), Workshop on Transferring and Adapting Source Knowledge in Computer Vision (TASK-CV) – year: 2016 ident: bib0170 article-title: Towards lifelong self-supervision: a deep learning direction for robotics publication-title: arXiv Preprint arXiv:1611.00201 – volume: 7 start-page: 123 year: 1995 end-page: 146 ident: bib0132 article-title: Catastrophic forgetting, rehearsal and pseudorehearsal publication-title: Connect. Sci. – volume: 87 start-page: 487 year: 2018 end-page: 504 ident: bib0078 article-title: Curiosity driven exploration of learned disentangled goal spaces publication-title: Proceedings of The 2nd Conference on Robot Learning – year: 2018 ident: bib0111 article-title: Variational continual learning publication-title: International Conference on Learning Representations – volume: 24 start-page: 109 year: 1989 end-page: 165 ident: bib0102 article-title: Catastrophic interference in connectionist networks: the sequential learning problem publication-title: Psychology of learning and motivation – start-page: 7765 year: 2018 end-page: 7773 ident: bib0096 article-title: Packnet: adding multiple tasks to a single network by iterative pruning publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition – volume: abs/1604.07316 year: 2016 ident: bib0010 article-title: End to end learning for self-driving cars publication-title: CoRR – volume: 47 start-page: 253 year: 2013 end-page: 279 ident: bib0008 article-title: The arcade learning environment: an evaluation platform for general agents publication-title: J. Artif. Intell. Res. – volume: 179 start-page: 41 year: 2018 end-page: 65 ident: bib0011 article-title: Pros and cons of gan evaluation measures publication-title: Comput. Vis. Image Understand. – volume: 23 start-page: 895 year: 2009 end-page: 896 ident: bib0032 article-title: Science of memory: concepts. henry l. roediger iii, yadin dudai, and susan m. fitzpatrick (eds.). oxford university press, new york, 2007. no. of pages 464. isbn 978-0-19-531044-3.(paperback) publication-title: Appl. Cognit. Psychol. – year: 2018 ident: bib0018 article-title: From babies to robots: the contribution of developmental robotics to developmental psychology publication-title: Child Dev. Perspect. – year: 2009 ident: bib0147 article-title: Active Learning Literature Survey publication-title: Technical Report – volume: abs/1901.00811 year: 2019 ident: bib0072 article-title: From exploration to control: learning object manipulation skills through novelty search and local adaptation publication-title: CoRR – start-page: 1408 year: 2014 end-page: 1413 ident: bib0155 article-title: Simultaneous on-line discovery and improvement of robotic skill options publication-title: Intelligent Robots and Systems (IROS 2014), 2014 IEEE/RSJ International Conference on – start-page: 2051 year: 2017 end-page: 2060 ident: bib0035 article-title: Multi-task self-supervised visual learning publication-title: Proceedings of the IEEE International Conference on Computer Vision – volume: abs/1907.05855 year: 2019 ident: bib0160 article-title: Discorl: continual reinforcement learning via policy distillation publication-title: CoRR – start-page: 5962 year: 2018 end-page: 5972 ident: bib0171 article-title: Memory replay gans: Learning to generate new categories without forgetting publication-title: Advances in Neural Information Processing Systems 31 – year: 2018 ident: bib0061 article-title: Quantitatively evaluating GANs with divergences proposed for training publication-title: International Conference on Learning Representations – year: 2018 ident: bib0040 article-title: Surreal: Open-source reinforcement learning framework and robot manipulation benchmark publication-title: Conference on Robot Learning – year: 2017 ident: bib0004 article-title: A multimodal dataset for object model learning from natural human-robot interaction publication-title: 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2017) – start-page: 3928 year: 2016 end-page: 3934 ident: bib0164 article-title: Stable reinforcement learning with autoencoders for tactile and visual data publication-title: 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) – volume: 39 start-page: 407 year: 2015 end-page: 428 ident: bib0064 article-title: Learning state representations with robotic priors publication-title: Autonomous Robot. – volume: 189 year: 2001 ident: bib0054 article-title: The scientist in the crib: minds, brains and how children learn publication-title: J. Nerv. Ment. Dis. – volume: 12 start-page: 1 year: 2017 end-page: 24 ident: bib0165 article-title: Diffusion-based neuromodulation can eliminate catastrophic forgetting in simple neural networks publication-title: PLOS ONE – start-page: 1540 year: 2015 end-page: 1548 ident: bib0119 article-title: Lifelong learning with non-iid tasks publication-title: Advances in Neural Information Processing Systems – year: 2017 ident: bib0069 article-title: Deep generative dual memory network for continual learning publication-title: ArXiv E-Prints – start-page: 1 year: 2019 end-page: 8 ident: bib0084 article-title: Deep unsupervised state representation learning with robotic priors: a robustness analysis publication-title: 2019 International Joint Conference on Neural Networks (IJCNN) – year: 2018 ident: bib0041 article-title: Towards robust evaluations of continual learning publication-title: arXiv preprint arXiv:1805.09733, Lifelong Learning: A Reinforcement Learning Approach (LLARLA) Workshop at FAIM 2018 – year: 2018 ident: bib0057 article-title: Overcoming catastrophic interference using conceptor-aided backpropagation publication-title: International Conference on Learning Representations – year: 2018 ident: bib0098 article-title: Roboturk: A crowdsourcing platform for robotic skill learning through imitation publication-title: Conference on Robot Learning – start-page: 1 year: 2019 end-page: 15 ident: bib0134 article-title: Derivative-free online learning of inverse dynamics models publication-title: IEEE Trans. Control Syst. Technol. – volume: 72 start-page: 2134 year: 2009 end-page: 2145 ident: bib0005 article-title: Using promoters and functional introns in genetic algorithms for neuroevolutionary learning in non-stationary problems publication-title: Neurocomputing – year: 2015 ident: bib0107 article-title: Never-ending learning publication-title: AAAI – start-page: 4904 year: 2016 end-page: 4911 ident: bib0117 article-title: Object identification from few examples by improving the invariance of a deep convolutional neural network publication-title: 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) – start-page: 2945 year: 2016 end-page: 2950 ident: bib0133 article-title: Online semi-parametric learning for inverse dynamics modeling publication-title: 2016 IEEE 55th Conference on Decision and Control (CDC) – year: 2018 ident: bib0036 article-title: Open-ended learning: a conceptual framework based on representational redescription publication-title: Front. Neurorobot. – year: 2016 ident: bib0137 article-title: Progressive neural networks publication-title: ArXiv E-Prints – year: 2019 ident: bib0120 article-title: A comprehensive, application-oriented study of catastrophic forgetting in DNNs publication-title: International Conference on Learning Representations – volume: 2 start-page: 230 year: 2010 end-page: 247 ident: bib0142 article-title: Formal theory of creativity, fun, and intrinsic motivation (1990–2010) publication-title: IEEE Trans. Auton. Ment. Dev. – volume: 12 start-page: 1 year: 2018 end-page: 207 ident: bib0024 article-title: Lifelong machine learning publication-title: Synthesis Lect. Artif. Intell. Mach. Learn. – volume: abs/1610.04286 year: 2016 ident: bib0138 article-title: Sim-to-real robot learning from pixels with progressive nets publication-title: CoRR – start-page: 3207 year: 2017 end-page: 3214 ident: bib0016 article-title: Incremental robot learning of new objects with fixed update time publication-title: 2017 IEEE International Conference on Robotics and Automation (ICRA) – start-page: 9769 year: 2018 end-page: 9776 ident: bib0055 article-title: Memory efficient experience replay for streaming learning publication-title: 2019 Int. Conf. Robot. Autom. (ICRA) – start-page: 3332 year: 2019 end-page: 3338 ident: bib0130 article-title: Closed-loop memory gan for continual learning publication-title: Proceedings of the 28th International Joint Conference on Artificial Intelligence – volume: 2 start-page: 340 year: 2010 end-page: 354 ident: bib0006 article-title: Multilevel darwinist brain (mdb): artificial evolution in a cognitive architecture for real robots – volume: 32 start-page: 53 year: 2015 end-page: 69 ident: bib0118 article-title: Visual domain adaptation: a survey of recent advances publication-title: IEEE Signal Process. Mag. – start-page: 3366 year: 2017 end-page: 3375 ident: bib0003 article-title: Expert gate: lifelong learning with a network of experts publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition – volume: 8 start-page: 924 year: 2016 end-page: 934 ident: bib0051 article-title: A bio-Inspired incremental learning architecture for applied perceptual problems publication-title: Cognit. Comput. – year: 2015 ident: bib0136 article-title: Policy Distillation publication-title: arXiv e-prints – start-page: 3738 year: 2018 end-page: 3748 ident: bib0131 article-title: Online structured laplace approximations for overcoming catastrophic forgetting publication-title: Advances in Neural Information Processing Systems 31 – volume: 28 start-page: 594 year: 2006 end-page: 611 ident: bib0042 article-title: One-shot learning of object categories publication-title: IEEE Trans. Pattern Anal. Mach. Intell. – start-page: 2672 year: 2014 end-page: 2680 ident: bib0052 article-title: Generative adversarial nets publication-title: Advances in Neural Information Processing Systems 27 – volume: 4 start-page: 504 year: 2013 ident: bib0103 article-title: The stability-plasticity dilemma: investigating the continuum from catastrophic forgetting to age-limited learning effects. publication-title: Front. Psychol. – start-page: 3420 year: 2017 end-page: 3429 ident: bib0151 article-title: Incremental learning of object detectors without catastrophic forgetting publication-title: IEEE International Conference on Computer Vision, ICCV 2017, Venice, Italy, October 22–29, 2017 – year: 2017 ident: bib0087 article-title: Learning without forgetting publication-title: IEEE Trans. Pattern Anal. Mach. Intell. – year: 2016 ident: bib0048 article-title: Active long term memory networks publication-title: ArXiv E-Prints – volume: 49 start-page: 8 year: 1961 end-page: 30 ident: bib0106 article-title: Steps toward artificial intelligence publication-title: Proc. IRE – year: 2008 ident: bib0012 article-title: Metalearning: Applications to Data Mining – start-page: 41 year: 2011 end-page: 51 ident: bib0156 article-title: Planning to be surprised: optimal bayesian exploration in dynamic environments publication-title: International Conference on Artificial General Intelligence – volume: 11 start-page: 13 year: 2005 end-page: 29 ident: bib0152 article-title: The development of embodied cognition: six lessons from babies publication-title: Artif. Life – year: 2005 ident: bib0129 article-title: Toward a formal framework for continual learning publication-title: NIPS workshop on Inductive Transfer, Whistler, Canada. – volume: 102 start-page: 419 year: 1995 ident: bib0101 article-title: Why there are complementary learning systems in the hippocampus and neocortex: insights from the successes and failures of connectionist models of learning and memory. publication-title: Psychol. Rev. – reference: , 2019. – volume: 80 start-page: 4528 year: 2018 end-page: 4537 ident: bib0144 article-title: Progress & compress: A scalable framework for continual learning publication-title: Proceedings of the 35th International Conference on Machine Learning – start-page: 2070 year: 2018 end-page: 2075 ident: bib0063 article-title: End-to-end race driving with deep reinforcement learning publication-title: 2018 IEEE Int. Conf. Robot. Automat. (ICRA) – volume: 5 start-page: 3 year: 2010 ident: bib0019 article-title: Toward an architecture for never-ending language learning. publication-title: AAAI – volume: abs/1810.06544 year: 2018 ident: bib0126 article-title: Deep imitative models for flexible inference, planning, and control publication-title: CoRR – start-page: 165 year: 1995 end-page: 196 ident: bib0159 article-title: Lifelong robot learning publication-title: The biology and technology of intelligent autonomous agents – year: 2019 ident: bib0020 article-title: S-TRIGGER: Continual State Representation Learning via Self-Triggered Generative Replay publication-title: arXiv e-prints – volume: 33 year: 2011 ident: bib0076 article-title: One shot learning of simple visual concepts publication-title: Proceedings of the Annual Meeting of the Cognitive Science Society – start-page: 588 year: 2016 end-page: 605 ident: bib0066 article-title: Fine-tuning deep neural networks in continuous learning scenarios publication-title: Asian Conference on Computer Vision – year: 2011 ident: bib0014 article-title: Statistics for High-Dimensional Data: Methods, Theory and Applications – start-page: 512 year: 2015 end-page: 519 ident: bib0045 article-title: Deep spatial autoencoders for visuomotor learning publication-title: 2016 IEEE Int. Conf. Robot. Automat. (ICRA) – volume: 111 start-page: 98 year: 2015 end-page: 136 ident: bib0039 article-title: The pascal visual object classes challenge: a retrospective publication-title: Int. J. Comput. Vis. – volume: 5 start-page: 12 year: 2018 ident: bib0013 article-title: Embodied evolution in collective robotics: a review publication-title: Front. Robot. AI – start-page: 1097 year: 2012 end-page: 1105 ident: bib0075 article-title: Imagenet classification with deep convolutional neural networks publication-title: Advances in neural information processing systems – start-page: 204 year: 1993 end-page: 211 ident: bib0122 article-title: Discriminability-based transfer between neural networks publication-title: Advances in neural information processing systems – start-page: 2234 year: 2016 end-page: 2242 ident: bib0139 article-title: Improved techniques for training gans publication-title: Advances in Neural Information Processing Systems 29 – start-page: 1 year: 2017 end-page: 9 ident: bib0025 article-title: End-to-end driving via conditional imitation learning publication-title: 2018 IEEE Int. Conf. Robot. Automat. (ICRA) – volume: abs/1607.00122 year: 2016 ident: bib0065 article-title: Less-forgetting learning in deep neural networks publication-title: CoRR – start-page: 21 year: 2015 end-page: 25 ident: bib0116 article-title: Teaching iCub to recognize objects using deep convolutional neural networks publication-title: Proceedings of the 4th International Conference on Machine Learning for Interactive Systems - Volume 43 – volume: 3 start-page: 128 year: 1999 end-page: 135 ident: bib0047 article-title: Catastrophic forgetting in connectionist networks publication-title: Trend. Cognit. Sci. – volume: 70 start-page: 3987 year: 2017 end-page: 3995 ident: bib0176 article-title: Continual learning through synaptic intelligence publication-title: Proceedings of the 34th International Conference on Machine Learning – start-page: 23 year: 2009 end-page: 65 ident: bib0162 article-title: Computing machinery and intelligence publication-title: Parsing the Turing Test – year: 2017 ident: bib0073 article-title: Overcoming catastrophic forgetting in neural networks publication-title: Proc. Nat. Acad. Sci. – year: 2009 ident: bib0148 article-title: Active Learning Literature Survey publication-title: Computer Sciences Technical Report – volume: 80 start-page: 1039 year: 2018 end-page: 1048 ident: bib0027 article-title: GEP-PG: decoupling exploration and exploitation in deep reinforcement learning algorithms publication-title: Proceedings of the 35th International Conference on Machine Learning – volume: 15 start-page: 1 year: 2019 end-page: 41 ident: bib0109 article-title: Learning and forgetting using reinforced bayesian change detection publication-title: PLOS Comput. Biol. – year: 2017 ident: bib0038 article-title: Learning State Representations for Robotic Control publication-title: M. Thesis – start-page: 13 year: 2015 end-page: 18 ident: bib0029 article-title: Exploration strategies for incremental learning of object-based visual saliency publication-title: 2015 Joint IEEE International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob) – start-page: 23 year: 2015 ident: bib0094 article-title: From passive to interactive object learning and recognition through self-identification on a humanoid robot publication-title: Auton. Robot. – reference: G. Csurka, A Comprehensive Survey on Domain Adaptation for Visual Applications, Springer International Publishing, Cham, pp. 1–35. doi: – year: 2017 ident: bib0046 article-title: Intrinsically motivated goal exploration processes with automatic curriculum learning publication-title: arXiv preprint arXiv:1708.02190 – volume: 518 start-page: 529 year: 2015 end-page: 533 ident: bib0108 article-title: Human-level control through deep reinforcement learning publication-title: Nature – year: 2019 ident: bib0089 article-title: Continual reinforcement learning in 3D non-stationary environments publication-title: arXiv preprint arXiv:1905.10112 – start-page: 6467 year: 2017 end-page: 6476 ident: bib0092 article-title: Gradient episodic memory for continual learning publication-title: Advances in Neural Information Processing Systems 30 – start-page: 526 year: 2017 end-page: 533 ident: bib0037 article-title: Neurogenesis deep learning: extending deep networks to accommodate new classes publication-title: 2017 Int. Joint Conf. Neural Netw. (IJCNN) – year: 2018 ident: bib0154 article-title: Memory-based parameter adaptation publication-title: International Conference on Learning Representations – reference: U. Michieli, P. Zanuttigh, Knowledge distillation for incremental learning in semantic segmentation, Provided by the SAO/NASA Astrophysics Data System. – year: 2019 ident: bib0023 article-title: Efficient lifelong learning with A-GEM publication-title: ICLR – volume: 112 start-page: 244 year: 2019 end-page: 259 ident: bib0030 article-title: Exploring to learn visual saliency: the rl-iac approach publication-title: Robot. Autonomous Syst. – start-page: 175 year: 2016 end-page: 184 ident: bib0090 article-title: Comparing incremental learning strategies for convolutional neural networks publication-title: IAPR Workshop on Artificial Neural Networks in Pattern Recognition – reference: F. Bellas, A. Faiña, G. Varela, R.J. Duro, A cognitive developmental robotics architecture for lifelong learning by evolution in real robots. The 2010 International Joint Conference on Neural Networks (IJCNN). IEEE, 2010. – volume: 18 start-page: 1 year: 2017 end-page: 43 ident: bib0062 article-title: Using conceptors to manage neural long-term memories for temporal patterns publication-title: J. Mach. Learn. Res. – year: 1994 ident: bib0128 publication-title: Continual Learning in Reinforcement Environments – volume: 78 start-page: 17 year: 2017 end-page: 26 ident: bib0091 article-title: CORe50: a New Dataset and Benchmark for Continuous Object Recognition publication-title: Proceedings of the 1st Annual Conference on Robot Learning – year: 2019 ident: bib0123 article-title: Decoupling feature extraction from policy learning: assessing benefits of state representation learning in goal based robotics publication-title: ICLR – volume: abs/1506.03365 year: 2015 ident: bib0175 article-title: Lsun: construction of a large-scale image dataset using deep learning with humans in the loop. publication-title: CoRR – year: 2013 ident: bib0053 article-title: An empirical investigation of catastrophic forgetting in gradient-Based neural networks publication-title: ArXiv E-Prints – start-page: 6626 year: 2017 end-page: 6637 ident: bib0059 article-title: Gans trained by a two time-scale update rule converge to a local nash equilibrium publication-title: Advances in Neural Information Processing Systems 30 – year: 2019 ident: bib0015 article-title: Large-scale study of curiosity-driven learning publication-title: ICLR – year: 2019 ident: bib0081 article-title: Generative Models from the perspective of Continual Learning publication-title: IJCNN - International Joint Conference on Neural Networks – year: 2017 ident: bib0124 article-title: Lifelong generative modeling publication-title: arXiv preprint arXiv:1705.09847 – volume: 34 start-page: 3 year: 2015 end-page: 25 ident: bib0028 article-title: Herbdisc: towards lifelong robotic object discovery publication-title: Int. J. Robot. Res. – start-page: 1 year: 2015 end-page: 9 ident: bib0060 article-title: Distilling the knowledge in a neural network publication-title: NIPS Deep Learning and Representation Learning Workshop (2015) – volume: 15 start-page: 151 year: 2003 end-page: 190 ident: bib0093 article-title: Developmental robotics: a survey publication-title: Connect. Sci. – volume: 113 start-page: 54 year: 2019 end-page: 71 ident: bib0114 article-title: Continual lifelong learning with neural networks: a review publication-title: Neural Netw. – start-page: 2990 year: 2017 end-page: 2999 ident: bib0150 article-title: Continual learning with deep generative replay publication-title: Advances in Neural Information Processing Systems – year: 2011 ident: bib0110 article-title: Reading digits in natural images with unsupervised feature learning publication-title: NIPS Workshop on Deep Learning and Unsupervised Feature Learning 2011 – year: 2019 ident: bib0033 article-title: Learning without memorizing publication-title: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) – start-page: 126 year: 2012 end-page: 133 ident: bib0100 article-title: Learn to swing up and balance a real pole based on raw visual input data publication-title: Neural Information Processing - 19th International Conference, ICONIP 2012, Doha, Qatar, November 12–15, 2012, Proceedings, Part V – year: 2017 ident: bib0071 article-title: Measuring catastrophic forgetting in neural networks publication-title: AAAI – year: 2017 ident: bib0143 article-title: Proximal policy optimization algorithms publication-title: arXiv preprint arXiv:1707.06347 – volume: 350 start-page: 1332 year: 2015 end-page: 1338 ident: bib0077 article-title: Human-level concept learning through probabilistic program induction publication-title: Science – volume: 97 start-page: 1331 year: 2019 end-page: 1340 ident: bib0026 article-title: CURIOUS: intrinsically motivated modular multi-goal reinforcement learning publication-title: Proceedings of the 36th International Conference on Machine Learning – year: 2017 ident: bib0161 article-title: Encoder based lifelong learning publication-title: IEEE Int. Conf. Comput. Vis. – start-page: 436 year: 1984 end-page: 445 ident: bib0163 article-title: A theory of the learnable publication-title: Proceedings of the sixteenth annual ACM symposium on Theory of computing – year: 2018 ident: bib0174 article-title: Lifelong learning with dynamically expandable networks publication-title: International Conference on Learning Representations – volume: abs/1705.08395 year: 2017 ident: bib0145 article-title: Continual learning in generative adversarial nets publication-title: CoRR – volume: 11 start-page: 265 year: 2007 end-page: 286 ident: bib0113 article-title: Intrinsic motivation systems for autonomous mental development publication-title: Evolut. Comput. IEEE Trans. – year: 2016 ident: bib0050 article-title: Incremental learning algorithms and applications publication-title: European Symposium on Artificial Neural Networks (ESANN) – year: 2016 ident: bib0149 article-title: Loss is its own reward: self-supervision for reinforcement learning publication-title: arXiv Preprint arXiv:1612.07307 – start-page: 2471 year: 2017 end-page: 2480 ident: bib0167 article-title: Growing a brain: fine-tuning by increasing model capacity publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition – year: 2019 ident: bib0068 article-title: Continual reinforcement learning deployed in real-life using policy distillation and sim2real transfer publication-title: ICML Workshop on Multi-Task and Lifelong Learning – volume: 12 start-page: 78 year: 2018 ident: bib0115 article-title: Lifelong learning of spatiotemporal representations with dual-memory recurrent self-organization publication-title: Front. Neurorobot. – start-page: 3462 year: 2017 end-page: 3468 ident: bib0177 article-title: Tensor based knowledge transfer across skill categories for robot control publication-title: Proceedings of the 26th International Joint Conference on Artificial Intelligence – reference: . – volume: abs/1802.10546 year: 2018 ident: bib0112 article-title: Computational theories of curiosity-driven learning publication-title: CoRR – year: 2019 ident: bib0058 article-title: NICO: A Dataset Towards Non-I.I.D. Image Classification publication-title: arXiv e-prints – year: 2015 ident: bib0141 article-title: Prioritized experience replay publication-title: arXiv preprint arXiv:1511.05952 – start-page: 3948 year: 2017 end-page: 3955 ident: bib0049 article-title: Learning to fly by crashing publication-title: 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) – start-page: 9895 year: 2018 end-page: 9905 ident: bib0001 article-title: Life-long disentangled representation learning with cross-domain latent homologies publication-title: Proceedings of the 32Nd International Conference on Neural Information Processing Systems – start-page: 4457 year: 2017 end-page: 4466 ident: bib0021 article-title: On-the-fly adaptation of regression forests for online camera relocalisation publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition – start-page: 544 year: 2016 end-page: 550 ident: bib0017 article-title: Incremental semiparametric inverse dynamics learning publication-title: Robotics and Automation (ICRA), 2016 IEEE International Conference on – start-page: 5533 year: 2017 end-page: 5542 ident: bib0125 article-title: icarl: Incremental classifier and representation learning publication-title: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) – start-page: 1126 year: 2017 end-page: 1135 ident: bib0044 article-title: Model-agnostic meta-learning for fast adaptation of deep networks publication-title: Proceedings of the 34th International Conference on Machine Learning - Volume 70 – year: 2018 ident: bib0022 article-title: Riemannian walk for incremental learning: Understanding forgetting and intransigence publication-title: ECCV – volume: 87 start-page: 651 year: 2018 end-page: 673 ident: bib0067 article-title: Scalable deep reinforcement learning for vision-based robotic manipulation publication-title: Proceedings of The 2nd Conference on Robot Learning – start-page: 604 year: 2019 end-page: 619 ident: bib0085 article-title: Training discriminative models to evaluate generative ones publication-title: Artificial Neural Networks and Machine Learning - ICANN 2019: Image Processing - 28th International Conference on Artificial Neural Networks, Munich, Germany, September 17–19, 2019, Proceedings, Part III – start-page: 2112 year: 2018 end-page: 21123 ident: bib0056 article-title: New metrics and experimental paradigms for continual learning publication-title: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) – volume: 10 start-page: 13:1 year: 2019 end-page: 13:37 ident: bib0166 article-title: A survey of zero-shot learning: settings, methods, and applications publication-title: ACM Trans. Intell. Syst. Technol. – volume: 22 start-page: 1453 year: 2012 end-page: 1461 ident: bib0178 article-title: Online incremental feature learning with denoising autoencoders publication-title: Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics – year: 2015 ident: 10.1016/j.inffus.2019.12.004_sbref0104 article-title: Never-ending learning – start-page: 1 year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0025 article-title: End-to-end driving via conditional imitation learning publication-title: 2018 IEEE Int. Conf. Robot. Automat. (ICRA) – start-page: 588 year: 2016 ident: 10.1016/j.inffus.2019.12.004_bib0066 article-title: Fine-tuning deep neural networks in continuous learning scenarios – year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0018 article-title: From babies to robots: the contribution of developmental robotics to developmental psychology publication-title: Child Dev. Perspect. doi: 10.1111/cdep.12282 – volume: 12 start-page: 1 issue: 3 year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0024 article-title: Lifelong machine learning publication-title: Synthesis Lect. Artif. Intell. Mach. Learn. – start-page: 13 year: 2015 ident: 10.1016/j.inffus.2019.12.004_bib0029 article-title: Exploration strategies for incremental learning of object-based visual saliency – start-page: 2945 year: 2016 ident: 10.1016/j.inffus.2019.12.004_bib0133 article-title: Online semi-parametric learning for inverse dynamics modeling – start-page: 2112 year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0056 article-title: New metrics and experimental paradigms for continual learning – volume: 8 start-page: 293 issue: 3–4 year: 1992 ident: 10.1016/j.inffus.2019.12.004_bib0088 article-title: Self-improving reactive agents based on reinforcement learning, planning and teaching publication-title: Mach. Learn. doi: 10.1023/A:1022628806385 – year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0111 article-title: Variational continual learning – start-page: 1408 year: 2014 ident: 10.1016/j.inffus.2019.12.004_bib0155 article-title: Simultaneous on-line discovery and improvement of robotic skill options – year: 2008 ident: 10.1016/j.inffus.2019.12.004_bib0012 – start-page: 23 year: 2015 ident: 10.1016/j.inffus.2019.12.004_bib0094 article-title: From passive to interactive object learning and recognition through self-identification on a humanoid robot publication-title: Auton. Robot. – volume: abs/1610.04286 year: 2016 ident: 10.1016/j.inffus.2019.12.004_bib0138 article-title: Sim-to-real robot learning from pixels with progressive nets publication-title: CoRR – start-page: 1126 year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0044 article-title: Model-agnostic meta-learning for fast adaptation of deep networks – start-page: 2051 year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0035 article-title: Multi-task self-supervised visual learning – start-page: 2070 year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0063 article-title: End-to-end race driving with deep reinforcement learning publication-title: 2018 IEEE Int. Conf. Robot. Automat. (ICRA) doi: 10.1109/ICRA.2018.8460934 – start-page: 204 year: 1993 ident: 10.1016/j.inffus.2019.12.004_bib0122 article-title: Discriminability-based transfer between neural networks – volume: 518 start-page: 529 issue: 7540 year: 2015 ident: 10.1016/j.inffus.2019.12.004_bib0108 article-title: Human-level control through deep reinforcement learning publication-title: Nature doi: 10.1038/nature14236 – start-page: 5533 year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0125 article-title: icarl: Incremental classifier and representation learning – start-page: 526 year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0037 article-title: Neurogenesis deep learning: extending deep networks to accommodate new classes publication-title: 2017 Int. Joint Conf. Neural Netw. (IJCNN) doi: 10.1109/IJCNN.2017.7965898 – volume: 37 start-page: 405 issue: 4–5 year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0158 article-title: The limits and potentials of deep learning for robotics publication-title: Int. J. Robot. Res. doi: 10.1177/0278364918770733 – year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0036 article-title: Open-ended learning: a conceptual framework based on representational redescription publication-title: Front. Neurorobot. doi: 10.3389/fnbot.2018.00059 – volume: 4 start-page: 504 issue: August year: 2013 ident: 10.1016/j.inffus.2019.12.004_bib0103 article-title: The stability-plasticity dilemma: investigating the continuum from catastrophic forgetting to age-limited learning effects. publication-title: Front. Psychol. – volume: 47 start-page: 253 year: 2013 ident: 10.1016/j.inffus.2019.12.004_bib0008 article-title: The arcade learning environment: an evaluation platform for general agents publication-title: J. Artif. Intell. Res. doi: 10.1613/jair.3912 – volume: 24 start-page: 109 year: 1989 ident: 10.1016/j.inffus.2019.12.004_bib0102 article-title: Catastrophic interference in connectionist networks: the sequential learning problem doi: 10.1016/S0079-7421(08)60536-8 – start-page: 23 year: 2009 ident: 10.1016/j.inffus.2019.12.004_bib0162 article-title: Computing machinery and intelligence – year: 2017 ident: 10.1016/j.inffus.2019.12.004_sbref0004 article-title: A multimodal dataset for object model learning from natural human-robot interaction – volume: abs/1901.00811 year: 2019 ident: 10.1016/j.inffus.2019.12.004_bib0072 article-title: From exploration to control: learning object manipulation skills through novelty search and local adaptation publication-title: CoRR – year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0057 article-title: Overcoming catastrophic interference using conceptor-aided backpropagation – year: 2009 ident: 10.1016/j.inffus.2019.12.004_bib0147 article-title: Active Learning Literature Survey – start-page: 175 year: 2016 ident: 10.1016/j.inffus.2019.12.004_bib0090 article-title: Comparing incremental learning strategies for convolutional neural networks – volume: 2 year: 1998 ident: 10.1016/j.inffus.2019.12.004_bib0157 – year: 2019 ident: 10.1016/j.inffus.2019.12.004_bib0068 article-title: Continual reinforcement learning deployed in real-life using policy distillation and sim2real transfer – year: 2019 ident: 10.1016/j.inffus.2019.12.004_sbref0079 article-title: Generative Models from the perspective of Continual Learning – year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0040 article-title: Surreal: Open-source reinforcement learning framework and robot manipulation benchmark – year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0073 article-title: Overcoming catastrophic forgetting in neural networks publication-title: Proc. Nat. Acad. Sci. doi: 10.1073/pnas.1611835114 – start-page: 7765 year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0096 article-title: Packnet: adding multiple tasks to a single network by iterative pruning – start-page: 1540 year: 2015 ident: 10.1016/j.inffus.2019.12.004_bib0119 article-title: Lifelong learning with non-iid tasks – volume: 72 start-page: 2134 issue: 10–12 year: 2009 ident: 10.1016/j.inffus.2019.12.004_bib0005 article-title: Using promoters and functional introns in genetic algorithms for neuroevolutionary learning in non-stationary problems publication-title: Neurocomputing doi: 10.1016/j.neucom.2008.10.022 – start-page: 3738 year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0131 article-title: Online structured laplace approximations for overcoming catastrophic forgetting – year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0161 article-title: Encoder based lifelong learning publication-title: IEEE Int. Conf. Comput. Vis. – volume: 350 start-page: 1332 issue: 6266 year: 2015 ident: 10.1016/j.inffus.2019.12.004_bib0077 article-title: Human-level concept learning through probabilistic program induction publication-title: Science doi: 10.1126/science.aab3050 – year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0041 article-title: Towards robust evaluations of continual learning publication-title: arXiv preprint arXiv:1805.09733, Lifelong Learning: A Reinforcement Learning Approach (LLARLA) Workshop at FAIM 2018 – year: 2010 ident: 10.1016/j.inffus.2019.12.004_bib0079 article-title: MNIST handwritten digit database publication-title: Public – ident: 10.1016/j.inffus.2019.12.004_bib0031 doi: 10.1007/978-3-319-58347-1_1 – volume: abs/1705.08395 year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0145 article-title: Continual learning in generative adversarial nets publication-title: CoRR – volume: 80 start-page: 4528 year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0144 article-title: Progress & compress: A scalable framework for continual learning – volume: 10 start-page: 13:1 issue: 2 year: 2019 ident: 10.1016/j.inffus.2019.12.004_bib0166 article-title: A survey of zero-shot learning: settings, methods, and applications publication-title: ACM Trans. Intell. Syst. Technol. doi: 10.1145/3293318 – volume: 12 start-page: 1 issue: 11 year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0165 article-title: Diffusion-based neuromodulation can eliminate catastrophic forgetting in simple neural networks publication-title: PLOS ONE doi: 10.1371/journal.pone.0187736 – volume: 189 year: 2001 ident: 10.1016/j.inffus.2019.12.004_bib0054 article-title: The scientist in the crib: minds, brains and how children learn publication-title: J. Nerv. Ment. Dis. – volume: 49 start-page: 8 issue: 1 year: 1961 ident: 10.1016/j.inffus.2019.12.004_bib0106 article-title: Steps toward artificial intelligence publication-title: Proc. IRE doi: 10.1109/JRPROC.1961.287775 – start-page: 3462 year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0177 article-title: Tensor based knowledge transfer across skill categories for robot control – start-page: 3366 year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0003 article-title: Expert gate: lifelong learning with a network of experts – volume: 15 start-page: 151 issue: 4 year: 2003 ident: 10.1016/j.inffus.2019.12.004_bib0093 article-title: Developmental robotics: a survey publication-title: Connect. Sci. doi: 10.1080/09540090310001655110 – volume: 113 start-page: 54 year: 2019 ident: 10.1016/j.inffus.2019.12.004_bib0114 article-title: Continual lifelong learning with neural networks: a review publication-title: Neural Netw. doi: 10.1016/j.neunet.2019.01.012 – volume: 11 start-page: 265 issue: 2 year: 2007 ident: 10.1016/j.inffus.2019.12.004_bib0113 article-title: Intrinsic motivation systems for autonomous mental development publication-title: Evolut. Comput. IEEE Trans. doi: 10.1109/TEVC.2006.890271 – start-page: 3406 year: 2015 ident: 10.1016/j.inffus.2019.12.004_bib0121 article-title: Supersizing self-supervision: learning to grasp from 50k tries and 700 robot hours publication-title: 2016 IEEE Int. Conf. Robot. Automat. (ICRA) – volume: abs/1604.07316 year: 2016 ident: 10.1016/j.inffus.2019.12.004_bib0010 article-title: End to end learning for self-driving cars publication-title: CoRR – year: 2019 ident: 10.1016/j.inffus.2019.12.004_sbref0056 article-title: NICO: A Dataset Towards Non-I.I.D. Image Classification publication-title: arXiv e-prints – volume: 33 year: 2011 ident: 10.1016/j.inffus.2019.12.004_bib0076 article-title: One shot learning of simple visual concepts – start-page: 1097 year: 2012 ident: 10.1016/j.inffus.2019.12.004_bib0075 article-title: Imagenet classification with deep convolutional neural networks – volume: abs/1802.00853 year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0172 article-title: Incremental classifier learning with generative adversarial networks publication-title: CoRR – volume: abs/1506.03365 year: 2015 ident: 10.1016/j.inffus.2019.12.004_bib0175 article-title: Lsun: construction of a large-scale image dataset using deep learning with humans in the loop. publication-title: CoRR – volume: 34 start-page: 3 issue: 1 year: 2015 ident: 10.1016/j.inffus.2019.12.004_bib0028 article-title: Herbdisc: towards lifelong robotic object discovery publication-title: Int. J. Robot. Res. doi: 10.1177/0278364914546030 – ident: 10.1016/j.inffus.2019.12.004_bib0105 – volume: 3 start-page: 128 issue: 4 year: 1999 ident: 10.1016/j.inffus.2019.12.004_bib0047 article-title: Catastrophic forgetting in connectionist networks publication-title: Trend. Cognit. Sci. doi: 10.1016/S1364-6613(99)01294-2 – start-page: 67 year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0095 article-title: Piggyback: adapting a single network to multiple tasks by learning to mask weights – start-page: 1 year: 2015 ident: 10.1016/j.inffus.2019.12.004_bib0060 article-title: Distilling the knowledge in a neural network – year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0098 article-title: Roboturk: A crowdsourcing platform for robotic skill learning through imitation – year: 2009 ident: 10.1016/j.inffus.2019.12.004_bib0148 article-title: Active Learning Literature Survey – year: 2009 ident: 10.1016/j.inffus.2019.12.004_bib0135 – volume: 15 start-page: 1 issue: 4 year: 2019 ident: 10.1016/j.inffus.2019.12.004_bib0109 article-title: Learning and forgetting using reinforced bayesian change detection publication-title: PLOS Comput. Biol. doi: 10.1371/journal.pcbi.1006713 – volume: 7 start-page: 123 issue: 2 year: 1995 ident: 10.1016/j.inffus.2019.12.004_bib0132 article-title: Catastrophic forgetting, rehearsal and pseudorehearsal publication-title: Connect. Sci. doi: 10.1080/09540099550039318 – start-page: 6467 year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0092 article-title: Gradient episodic memory for continual learning – year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0154 article-title: Memory-based parameter adaptation – volume: 179 start-page: 41 year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0011 article-title: Pros and cons of gan evaluation measures publication-title: Comput. Vis. Image Understand. doi: 10.1016/j.cviu.2018.10.009 – year: 2019 ident: 10.1016/j.inffus.2019.12.004_bib0033 article-title: Learning without memorizing – year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0069 article-title: Deep generative dual memory network for continual learning publication-title: ArXiv E-Prints – year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0061 article-title: Quantitatively evaluating GANs with divergences proposed for training – volume: abs/1607.00122 year: 2016 ident: 10.1016/j.inffus.2019.12.004_bib0065 article-title: Less-forgetting learning in deep neural networks publication-title: CoRR – year: 2015 ident: 10.1016/j.inffus.2019.12.004_sbref0133 article-title: Policy Distillation publication-title: arXiv e-prints – start-page: 1 year: 2019 ident: 10.1016/j.inffus.2019.12.004_bib0084 article-title: Deep unsupervised state representation learning with robotic priors: a robustness analysis – start-page: 3420 year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0151 article-title: Incremental learning of object detectors without catastrophic forgetting – year: 1994 ident: 10.1016/j.inffus.2019.12.004_bib0128 – start-page: 9769 year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0055 article-title: Memory efficient experience replay for streaming learning publication-title: 2019 Int. Conf. Robot. Autom. (ICRA) – volume: 17 start-page: 1334 issue: 1 year: 2016 ident: 10.1016/j.inffus.2019.12.004_bib0086 article-title: End-to-end training of deep visuomotor policies publication-title: J. Mach. Learn. Res. – volume: 87 start-page: 487 year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0078 article-title: Curiosity driven exploration of learned disentangled goal spaces – start-page: 2234 year: 2016 ident: 10.1016/j.inffus.2019.12.004_bib0139 article-title: Improved techniques for training gans – year: 2011 ident: 10.1016/j.inffus.2019.12.004_bib0110 article-title: Reading digits in natural images with unsupervised feature learning – start-page: 3332 year: 2019 ident: 10.1016/j.inffus.2019.12.004_bib0130 article-title: Closed-loop memory gan for continual learning – volume: 80 start-page: 4548 year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0146 article-title: Overcoming catastrophic forgetting with hard attention to the task publication-title: ICML – volume: 78 start-page: 17 year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0091 article-title: CORe50: a New Dataset and Benchmark for Continuous Object Recognition – volume: 87 start-page: 651 year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0067 article-title: Scalable deep reinforcement learning for vision-based robotic manipulation – start-page: 1 year: 2019 ident: 10.1016/j.inffus.2019.12.004_bib0134 article-title: Derivative-free online learning of inverse dynamics models publication-title: IEEE Trans. Control Syst. Technol. – year: 2016 ident: 10.1016/j.inffus.2019.12.004_bib0137 article-title: Progressive neural networks publication-title: ArXiv E-Prints – year: 2011 ident: 10.1016/j.inffus.2019.12.004_bib0014 – year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0071 article-title: Measuring catastrophic forgetting in neural networks – year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0087 article-title: Learning without forgetting publication-title: IEEE Trans. Pattern Anal. Mach. Intell. – year: 2019 ident: 10.1016/j.inffus.2019.12.004_bib0120 article-title: A comprehensive, application-oriented study of catastrophic forgetting in DNNs – volume: abs/1810.06544 year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0126 article-title: Deep imitative models for flexible inference, planning, and control publication-title: CoRR – year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0174 article-title: Lifelong learning with dynamically expandable networks – volume: 23 start-page: 895 issue: 6 year: 2009 ident: 10.1016/j.inffus.2019.12.004_bib0032 article-title: Science of memory: concepts. henry l. roediger iii, yadin dudai, and susan m. fitzpatrick (eds.). oxford university press, new york, 2007. no. of pages 464. isbn 978-0-19-531044-3.(paperback) publication-title: Appl. Cognit. Psychol. doi: 10.1002/acp.1533 – year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0173 article-title: Fashion-mnist: a novel image dataset for benchmarking machine learning algorithms publication-title: arXiv preprint arXiv:1708.07747 – year: 2019 ident: 10.1016/j.inffus.2019.12.004_sbref0019 article-title: S-TRIGGER: Continual State Representation Learning via Self-Triggered Generative Replay publication-title: arXiv e-prints – year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0143 article-title: Proximal policy optimization algorithms publication-title: arXiv preprint arXiv:1707.06347 – year: 2013 ident: 10.1016/j.inffus.2019.12.004_bib0053 article-title: An empirical investigation of catastrophic forgetting in gradient-Based neural networks publication-title: ArXiv E-Prints – start-page: 41 year: 2009 ident: 10.1016/j.inffus.2019.12.004_bib0009 article-title: Curriculum learning – start-page: 3207 year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0016 article-title: Incremental robot learning of new objects with fixed update time – start-page: 3948 year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0049 article-title: Learning to fly by crashing – start-page: 2990 year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0150 article-title: Continual learning with deep generative replay – year: 2019 ident: 10.1016/j.inffus.2019.12.004_bib0104 article-title: Incremental learning techniques for semantic segmentation – volume: 2 start-page: 230 issue: 3 year: 2010 ident: 10.1016/j.inffus.2019.12.004_bib0142 article-title: Formal theory of creativity, fun, and intrinsic motivation (1990–2010) publication-title: IEEE Trans. Auton. Ment. Dev. doi: 10.1109/TAMD.2010.2056368 – volume: abs/1701.08734 year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0043 article-title: Pathnet: evolution channels gradient descent in super neural networks publication-title: CoRR – volume: 70 start-page: 3987 year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0176 article-title: Continual learning through synaptic intelligence – volume: 102 start-page: 419 issue: 3 year: 1995 ident: 10.1016/j.inffus.2019.12.004_bib0101 article-title: Why there are complementary learning systems in the hippocampus and neocortex: insights from the successes and failures of connectionist models of learning and memory. publication-title: Psychol. Rev. doi: 10.1037/0033-295X.102.3.419 – year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0038 article-title: Learning State Representations for Robotic Control publication-title: M. Thesis – volume: 97 start-page: 1331 year: 2019 ident: 10.1016/j.inffus.2019.12.004_bib0026 article-title: CURIOUS: intrinsically motivated modular multi-goal reinforcement learning – start-page: 436 year: 1984 ident: 10.1016/j.inffus.2019.12.004_bib0163 article-title: A theory of the learnable – start-page: 2471 year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0167 article-title: Growing a brain: fine-tuning by increasing model capacity – volume: 5 start-page: 3 year: 2010 ident: 10.1016/j.inffus.2019.12.004_bib0019 article-title: Toward an architecture for never-ending language learning. – start-page: 466 year: 2019 ident: 10.1016/j.inffus.2019.12.004_bib0083 article-title: Marginal replay vs conditional replay for continual learning – start-page: 4652 year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0080 article-title: Overcoming catastrophic forgetting by incremental moment matching – year: 2018 ident: 10.1016/j.inffus.2019.12.004_sbref0032 article-title: Don’t forget, there is more than forgetting: new metrics for Continual Learning – start-page: 6626 year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0059 article-title: Gans trained by a two time-scale update rule converge to a local nash equilibrium – year: 2010 ident: 10.1016/j.inffus.2019.12.004_bib0168 article-title: Caltech-UCSD Birds 200 – start-page: 5962 year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0171 article-title: Memory replay gans: Learning to generate new categories without forgetting – volume: 32 start-page: 53 issue: 3 year: 2015 ident: 10.1016/j.inffus.2019.12.004_bib0118 article-title: Visual domain adaptation: a survey of recent advances publication-title: IEEE Signal Process. Mag. doi: 10.1109/MSP.2014.2347059 – volume: 11 start-page: 13 year: 2005 ident: 10.1016/j.inffus.2019.12.004_bib0152 article-title: The development of embodied cognition: six lessons from babies publication-title: Artif. Life doi: 10.1162/1064546053278973 – volume: abs/1907.05855 year: 2019 ident: 10.1016/j.inffus.2019.12.004_bib0160 article-title: Discorl: continual reinforcement learning via policy distillation publication-title: CoRR – start-page: 5074 year: 2016 ident: 10.1016/j.inffus.2019.12.004_bib0002 article-title: Learning to poke by poking: experiential learning of intuitive physics – year: 2016 ident: 10.1016/j.inffus.2019.12.004_sbref0048 article-title: Incremental learning algorithms and applications – volume: 12 start-page: 78 year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0115 article-title: Lifelong learning of spatiotemporal representations with dual-memory recurrent self-organization publication-title: Front. Neurorobot. doi: 10.3389/fnbot.2018.00078 – year: 2005 ident: 10.1016/j.inffus.2019.12.004_bib0129 article-title: Toward a formal framework for continual learning – start-page: 4904 year: 2016 ident: 10.1016/j.inffus.2019.12.004_bib0117 article-title: Object identification from few examples by improving the invariance of a deep convolutional neural network – year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0070 article-title: Fearnet: brain-inspired model for incremental learning – year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0124 article-title: Lifelong generative modeling publication-title: arXiv preprint arXiv:1705.09847 – start-page: 41 year: 2011 ident: 10.1016/j.inffus.2019.12.004_bib0156 article-title: Planning to be surprised: optimal bayesian exploration in dynamic environments – volume: 28 start-page: 594 issue: 4 year: 2006 ident: 10.1016/j.inffus.2019.12.004_bib0042 article-title: One-shot learning of object categories publication-title: IEEE Trans. Pattern Anal. Mach. Intell. doi: 10.1109/TPAMI.2006.79 – start-page: 4457 year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0021 article-title: On-the-fly adaptation of regression forests for online camera relocalisation – volume: 39 start-page: 407 issue: 3 year: 2015 ident: 10.1016/j.inffus.2019.12.004_bib0064 article-title: Learning state representations with robotic priors publication-title: Autonomous Robot. doi: 10.1007/s10514-015-9459-7 – year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0046 article-title: Intrinsically motivated goal exploration processes with automatic curriculum learning publication-title: arXiv preprint arXiv:1708.02190 – year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0082 article-title: State representation learning for control: an overview publication-title: Neur. Netw. doi: 10.1016/j.neunet.2018.07.006 – volume: 291 start-page: 599 issue: 5504 year: 2001 ident: 10.1016/j.inffus.2019.12.004_bib0169 article-title: Autonomous mental development by robots and animals publication-title: Science doi: 10.1126/science.291.5504.599 – year: 2016 ident: 10.1016/j.inffus.2019.12.004_bib0048 article-title: Active long term memory networks publication-title: ArXiv E-Prints – volume: 10 start-page: 539 issue: 3 year: 1998 ident: 10.1016/j.inffus.2019.12.004_bib0140 article-title: A bayesian on-line change detection algorithm with process monitoring applications publication-title: Qual. Eng. doi: 10.1080/08982119808919168 – year: 2019 ident: 10.1016/j.inffus.2019.12.004_bib0127 article-title: Learning to learn without forgetting by maximizing transfer and minimizing interference – year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0022 article-title: Riemannian walk for incremental learning: Understanding forgetting and intransigence – volume: 22 start-page: 1453 year: 2012 ident: 10.1016/j.inffus.2019.12.004_bib0178 article-title: Online incremental feature learning with denoising autoencoders – year: 2009 ident: 10.1016/j.inffus.2019.12.004_bib0074 article-title: Learning multiple layers of features from tiny images – volume: abs/1802.10546 year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0112 article-title: Computational theories of curiosity-driven learning publication-title: CoRR – year: 2019 ident: 10.1016/j.inffus.2019.12.004_bib0123 article-title: Decoupling feature extraction from policy learning: assessing benefits of state representation learning in goal based robotics – volume: 8 start-page: 924 year: 2016 ident: 10.1016/j.inffus.2019.12.004_bib0051 article-title: A bio-Inspired incremental learning architecture for applied perceptual problems publication-title: Cognit. Comput. doi: 10.1007/s12559-016-9389-5 – start-page: 21 year: 2015 ident: 10.1016/j.inffus.2019.12.004_bib0116 article-title: Teaching iCub to recognize objects using deep convolutional neural networks – volume: 2 start-page: 340 year: 2010 ident: 10.1016/j.inffus.2019.12.004_bib0006 – start-page: 512 year: 2015 ident: 10.1016/j.inffus.2019.12.004_bib0045 article-title: Deep spatial autoencoders for visuomotor learning publication-title: 2016 IEEE Int. Conf. Robot. Automat. (ICRA) – ident: 10.1016/j.inffus.2019.12.004_bib0007 doi: 10.1109/IJCNN.2010.5596771 – volume: 5 start-page: 12 year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0013 article-title: Embodied evolution in collective robotics: a review publication-title: Front. Robot. AI doi: 10.3389/frobt.2018.00012 – start-page: 3928 year: 2016 ident: 10.1016/j.inffus.2019.12.004_bib0164 article-title: Stable reinforcement learning with autoencoders for tactile and visual data – year: 2019 ident: 10.1016/j.inffus.2019.12.004_bib0015 article-title: Large-scale study of curiosity-driven learning – volume: 116 start-page: 56 year: 2019 ident: 10.1016/j.inffus.2019.12.004_bib0097 article-title: Continuous learning in single-incremental-task scenarios publication-title: Neural Netw. doi: 10.1016/j.neunet.2019.03.010 – volume: 112 start-page: 244 year: 2019 ident: 10.1016/j.inffus.2019.12.004_bib0030 article-title: Exploring to learn visual saliency: the rl-iac approach publication-title: Robot. Autonomous Syst. doi: 10.1016/j.robot.2018.11.012 – year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0099 article-title: Unicorn: continual learning with a universal, off-policy agent publication-title: arXiv Preprint arXiv:1802.08294 – year: 2016 ident: 10.1016/j.inffus.2019.12.004_bib0149 article-title: Loss is its own reward: self-supervision for reinforcement learning publication-title: arXiv Preprint arXiv:1612.07307 – year: 2019 ident: 10.1016/j.inffus.2019.12.004_bib0023 article-title: Efficient lifelong learning with A-GEM – year: 2016 ident: 10.1016/j.inffus.2019.12.004_bib0170 article-title: Towards lifelong self-supervision: a deep learning direction for robotics publication-title: arXiv Preprint arXiv:1611.00201 – year: 2019 ident: 10.1016/j.inffus.2019.12.004_bib0089 article-title: Continual reinforcement learning in 3D non-stationary environments publication-title: arXiv preprint arXiv:1905.10112 – start-page: 604 year: 2019 ident: 10.1016/j.inffus.2019.12.004_bib0085 article-title: Training discriminative models to evaluate generative ones – start-page: 165 year: 1995 ident: 10.1016/j.inffus.2019.12.004_bib0159 article-title: Lifelong robot learning – start-page: 126 year: 2012 ident: 10.1016/j.inffus.2019.12.004_bib0100 article-title: Learn to swing up and balance a real pole based on raw visual input data – volume: 80 start-page: 1039 year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0027 article-title: GEP-PG: decoupling exploration and exploitation in deep reinforcement learning algorithms – start-page: 2672 year: 2014 ident: 10.1016/j.inffus.2019.12.004_bib0052 article-title: Generative adversarial nets – volume: abs/1811.07017 year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0153 article-title: On training recurrent neural networks for lifelong learning publication-title: CoRR – start-page: 9895 year: 2018 ident: 10.1016/j.inffus.2019.12.004_bib0001 article-title: Life-long disentangled representation learning with cross-domain latent homologies – volume: 18 start-page: 1 issue: 13 year: 2017 ident: 10.1016/j.inffus.2019.12.004_bib0062 article-title: Using conceptors to manage neural long-term memories for temporal patterns publication-title: J. Mach. Learn. Res. – volume: 111 start-page: 98 issue: 1 year: 2015 ident: 10.1016/j.inffus.2019.12.004_bib0039 article-title: The pascal visual object classes challenge: a retrospective publication-title: Int. J. Comput. Vis. doi: 10.1007/s11263-014-0733-5 – start-page: 544 year: 2016 ident: 10.1016/j.inffus.2019.12.004_bib0017 article-title: Incremental semiparametric inverse dynamics learning – year: 2015 ident: 10.1016/j.inffus.2019.12.004_bib0141 article-title: Prioritized experience replay publication-title: arXiv preprint arXiv:1511.05952 |
| SSID | ssj0017031 |
| Score | 2.6919448 |
| Snippet | •State of the art on continual / lifelong learning and its implications for robotics.•Proposal of a framework to present Continual Learning algorithms.•Summary... Continual learning (CL) is a particular machine learning paradigm where the data distribution and learning objective changes through time, or where all the... |
| SourceID | hal crossref elsevier |
| SourceType | Open Access Repository Enrichment Source Index Database Publisher |
| StartPage | 52 |
| SubjectTerms | Artificial Intelligence Catastrophic Forgetting Computational Engineering, Finance, and Science Computer Science Computer Vision and Pattern Recognition Continual Learning Deep Learning Lifelong Learning Reinforcement Learning Robotics |
| Title | Continual learning for robotics: Definition, framework, learning strategies, opportunities and challenges |
| URI | https://dx.doi.org/10.1016/j.inffus.2019.12.004 https://hal.science/hal-02381343 |
| Volume | 58 |
| WOSCitedRecordID | wos000516799200005&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVESC databaseName: Elsevier SD Freedom Collection Journals 2021 customDbUrl: eissn: 1872-6305 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0017031 issn: 1566-2535 databaseCode: AIEXJ dateStart: 20000701 isFulltext: true titleUrlDefault: https://www.sciencedirect.com providerName: Elsevier |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1JbxMxFLZCygEOLAVE2WQhjjHKeDwbt6iLUhhCBQX1ZsWOR0xVzUTNoor_w__keZ1EFWo5cBlFjp1M_L543nv-_D2E3iVaEy2fCZKoIiFMyIQI8AxIJPNZLBSLM7Oj-6PMJpP87Kw46fV--7Mw64usafKrq2L-X00NbWBsfXT2H8wdPhQa4DUYHa5gdrjeyvBab6o2ovmlT3toKuHXVrRaklmnAA5UVTeGq-V8V8vPMjG6H-Nla-0y8mWu_fRVY_RXzYbDvq_Cstj0b93pJktgXC02NvlLtXBF6y087A59gFXZwgTB6mx4tzUgsfnVhuzPsq2nTaBf1l0WXZfErgM1X22mMOiwo1rZvJpzAjaX4TQlNLFCJu-VbcszStLYHNK-vvDbHMS5jlbg12nKXmHSvLa28bbO9nj0jZ8cHPHyePJp-90NcuJ4VMIVppIYryZm8RqC7B2aJUXeRzuj48Ozj2G7ShcBMMK87rb9GU1DJLx-S3_zge789Nl8492cPkIPXFiCRxZOj1FPNbvooS_5gd0TYBfd_xxkfhdPUB2whj1uMNgfe6x9wB3SBjjgbND17lA2wFsYw4Ax3GHsKfp-dHi6PyaudgeRLKFLkotCRkOhqgpCOCUkLAlMSiZnCiJ4cHqnVaQyOc1YOqVKRTKVQ8aKiqZFNlVRLuJnqN-0jXqOMETsokrhQcOqGWOyEjQfVmLG0opSlSm5h2I_m1w6YXtdX-WCewbjObc24NoGPKIcbLCHSBg1t8IuN_TPvKG4c06t08kBgjeMfAuzFb5E67kDuLhu66D14jadXqJ73d_nFeovL1fqNbor18t6cfnGgfIPIwS81g |
| linkProvider | Elsevier |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Continual+Learning+for+Robotics%3A+Definition%2C+Framework%2C+Learning+Strategies%2C+Opportunities+and+Challenges&rft.jtitle=Information+fusion&rft.au=Lesort%2C+Timoth%C3%A9e&rft.au=Lomonaco%2C+Vincenzo&rft.au=Stoian%2C+Andrei&rft.au=Maltoni%2C+Davide&rft.date=2020-06-01&rft.pub=Elsevier&rft.issn=1566-2535&rft.eissn=1872-6305&rft_id=info:doi/10.1016%2Fj.inffus.2019.12.004&rft.externalDBID=HAS_PDF_LINK&rft.externalDocID=oai%3AHAL%3Ahal-02381343v1 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1566-2535&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1566-2535&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1566-2535&client=summon |