Class-incremental learning with causal relational replay

In Class-Incremental Learning (Class-IL), deep neural networks often fail to learn a sequence of classes incrementally due to catastrophic forgetting, a phenomenon arising from the absence of exposure to old knowledge. To alleviate this issue, conventional rehearsal methods, such as experience repla...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Expert systems with applications Ročník 250; s. 123901
Hlavní autori: Nguyen, Toan, Kieu, Duc, Duong, Bao, Kieu, Tung, Do, Kien, Nguyen, Thin, Le, Bac
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Elsevier Ltd 15.09.2024
Predmet:
ISSN:0957-4174
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract In Class-Incremental Learning (Class-IL), deep neural networks often fail to learn a sequence of classes incrementally due to catastrophic forgetting, a phenomenon arising from the absence of exposure to old knowledge. To alleviate this issue, conventional rehearsal methods, such as experience replay, store a limited number of old exemplars and then interleave with the current data for joint learning and rehearsal. However, the networks following this training scheme might not successfully reduce forgetting due to the lack of direct consideration of relations between samples of previously learned and new classes. Drawing inspiration from how humans learn by noticing the similarities and differences between classes, we propose a novel Class-IL framework called Relational Replay (RR). RR learns and recalls relations between images across all classes over time. To ensure these relations remain intrinsic and robust to forgetting, we incorporate causal reasoning to RR, resulting in Causal Relational Replay (CRR). CRR analyzes these relations using a causality perspective, aiming to identify intrinsic relations rooted in the images’ semantic features, serving as the cause of these relations. Our proposed method shows a competitive performance compared to the state-of-the-art rehearsal methods in Class-IL with clear and consistent improvements in the majority of settings on standard benchmark datasets. •We tackle the problem of catastrophic forgetting in Class-IL.•Our method leverages object relations, inspired by human memory, to prevent forgetting.•We discover spurious relations worsen forgetting, best to focus only on intrinsic ones.•We add a causal learning mechanism to analyze relations and combat spurious factors.•Our algorithm excels in Class-IL, consistently surpassing top rehearsal techniques.
AbstractList In Class-Incremental Learning (Class-IL), deep neural networks often fail to learn a sequence of classes incrementally due to catastrophic forgetting, a phenomenon arising from the absence of exposure to old knowledge. To alleviate this issue, conventional rehearsal methods, such as experience replay, store a limited number of old exemplars and then interleave with the current data for joint learning and rehearsal. However, the networks following this training scheme might not successfully reduce forgetting due to the lack of direct consideration of relations between samples of previously learned and new classes. Drawing inspiration from how humans learn by noticing the similarities and differences between classes, we propose a novel Class-IL framework called Relational Replay (RR). RR learns and recalls relations between images across all classes over time. To ensure these relations remain intrinsic and robust to forgetting, we incorporate causal reasoning to RR, resulting in Causal Relational Replay (CRR). CRR analyzes these relations using a causality perspective, aiming to identify intrinsic relations rooted in the images’ semantic features, serving as the cause of these relations. Our proposed method shows a competitive performance compared to the state-of-the-art rehearsal methods in Class-IL with clear and consistent improvements in the majority of settings on standard benchmark datasets. •We tackle the problem of catastrophic forgetting in Class-IL.•Our method leverages object relations, inspired by human memory, to prevent forgetting.•We discover spurious relations worsen forgetting, best to focus only on intrinsic ones.•We add a causal learning mechanism to analyze relations and combat spurious factors.•Our algorithm excels in Class-IL, consistently surpassing top rehearsal techniques.
ArticleNumber 123901
Author Duong, Bao
Nguyen, Thin
Nguyen, Toan
Kieu, Tung
Do, Kien
Le, Bac
Kieu, Duc
Author_xml – sequence: 1
  givenname: Toan
  orcidid: 0000-0003-2734-0622
  surname: Nguyen
  fullname: Nguyen, Toan
  email: s222165627@deakin.edu.au
  organization: Applied Artificial Intelligence Institute, Deakin University, Australia
– sequence: 2
  givenname: Duc
  orcidid: 0009-0008-4359-3383
  surname: Kieu
  fullname: Kieu, Duc
  email: 18127080@student.hcmus.edu.vn
  organization: Faculty of Information Technology, University of Science, Ho Chi Minh City, Viet Nam
– sequence: 3
  givenname: Bao
  orcidid: 0000-0001-9850-0270
  surname: Duong
  fullname: Duong, Bao
  email: duongng@deakin.edu.au
  organization: Applied Artificial Intelligence Institute, Deakin University, Australia
– sequence: 4
  givenname: Tung
  orcidid: 0000-0002-7696-1444
  surname: Kieu
  fullname: Kieu, Tung
  email: tungkvt@cs.aau.dk
  organization: Department of Computer Science, Aalborg University, Denmark
– sequence: 5
  givenname: Kien
  orcidid: 0000-0002-0119-122X
  surname: Do
  fullname: Do, Kien
  email: k.do@deakin.edu.au
  organization: Applied Artificial Intelligence Institute, Deakin University, Australia
– sequence: 6
  givenname: Thin
  orcidid: 0000-0003-3467-8963
  surname: Nguyen
  fullname: Nguyen, Thin
  email: thin.nguyen@deakin.edu.au
  organization: Applied Artificial Intelligence Institute, Deakin University, Australia
– sequence: 7
  givenname: Bac
  orcidid: 0000-0002-4306-6945
  surname: Le
  fullname: Le, Bac
  email: lhbac@fit.hcmus.edu.vn
  organization: Faculty of Information Technology, University of Science, Ho Chi Minh City, Viet Nam
BookMark eNp9j81KAzEUhbOoYFt9AVd9gRlvJpn8gBsp_kHBja5DJnNHU9JMSaKlb29rXbno6h4ufIfzzcgkjhEJuaFQU6Didl1j3tm6gYbXtGEa6IRMQbey4lTySzLLeQ1AJYCcErUMNufKR5dwg7HYsAhoU_TxY7Hz5XPh7Fc-PBMGW_wYf-M22P0VuRhsyHj9d-fk_fHhbflcrV6fXpb3q8oxgFIx1qlWWi2E1BwoqKGnrlND5wamFWe9EBb6TrsGEAelmhYF7y1VWkjeY8vmpDn1ujTmnHAw2-Q3Nu0NBXP0NWtz9DVHX3PyPUDqH-R8-d1fkvXhPHp3QvEg9e0xmew8Roe9T-iK6Ud_Dv8B1OV1PQ
CitedBy_id crossref_primary_10_1016_j_eswa_2024_125135
crossref_primary_10_1016_j_neunet_2024_106788
crossref_primary_10_1016_j_eswa_2025_129416
crossref_primary_10_1016_j_eswa_2025_128533
crossref_primary_10_1016_j_neucom_2025_129860
crossref_primary_10_1109_ACCESS_2025_3573041
crossref_primary_10_1016_j_eswa_2025_126399
crossref_primary_10_1016_j_eswa_2024_125057
Cites_doi 10.1037/0033-295X.97.2.285
10.2307/2331554
10.1111/rssb.12167
10.1016/j.eswa.2023.120044
10.1214/09-SS057
10.1145/3147.3165
10.1007/s10489-022-04045-7
10.1162/089976698300017197
10.1038/nature14539
10.1198/016214504000001880
10.1016/j.eswa.2022.118580
10.1145/3386252
10.1016/j.neunet.2019.03.010
10.1016/j.eswa.2022.118934
10.1016/j.neucom.2021.10.021
10.1371/journal.pone.0203848
10.1016/j.neunet.2019.01.012
ContentType Journal Article
Copyright 2024
Copyright_xml – notice: 2024
DBID AAYXX
CITATION
DOI 10.1016/j.eswa.2024.123901
DatabaseName CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Computer Science
ExternalDocumentID 10_1016_j_eswa_2024_123901
S095741742400767X
GroupedDBID --K
--M
.DC
.~1
0R~
13V
1B1
1RT
1~.
1~5
4.4
457
4G.
5GY
5VS
7-5
71M
8P~
9JN
9JO
AAAKF
AABNK
AACTN
AAEDT
AAEDW
AAIKJ
AAKOC
AALRI
AAOAW
AAQFI
AARIN
AAXUO
AAYFN
ABBOA
ABFNM
ABMAC
ABMVD
ABUCO
ACDAQ
ACGFS
ACHRH
ACNTT
ACRLP
ACZNC
ADBBV
ADEZE
ADTZH
AEBSH
AECPX
AEKER
AENEX
AFKWA
AFTJW
AGHFR
AGUBO
AGUMN
AGYEJ
AHHHB
AHJVU
AHZHX
AIALX
AIEXJ
AIKHN
AITUG
AJOXV
AKRWK
ALEQD
ALMA_UNASSIGNED_HOLDINGS
AMFUW
AMRAJ
AOUOD
APLSM
AXJTR
BJAXD
BKOJK
BLXMC
BNSAS
CS3
DU5
EBS
EFJIC
EO8
EO9
EP2
EP3
F5P
FDB
FIRID
FNPLU
FYGXN
G-Q
GBLVA
GBOLZ
HAMUX
IHE
J1W
JJJVA
KOM
LG9
LY1
LY7
M41
MO0
N9A
O-L
O9-
OAUVE
OZT
P-8
P-9
P2P
PC.
PQQKQ
Q38
RIG
ROL
RPZ
SDF
SDG
SDP
SDS
SES
SEW
SPC
SPCBC
SSB
SSD
SSL
SST
SSV
SSZ
T5K
TN5
~G-
29G
9DU
AAAKG
AAQXK
AATTM
AAXKI
AAYWO
AAYXX
ABJNI
ABKBG
ABUFD
ABWVN
ABXDB
ACLOT
ACNNM
ACRPL
ACVFH
ADCNI
ADJOM
ADMUD
ADNMO
AEIPS
AEUPX
AFJKZ
AFPUW
AGQPQ
AIGII
AIIUN
AKBMS
AKYEP
ANKPU
APXCP
ASPBG
AVWKF
AZFZN
CITATION
EFKBS
EFLBG
EJD
FEDTE
FGOYB
G-2
HLZ
HVGLF
HZ~
R2-
SBC
SET
WUQ
XPP
ZMT
~HD
ID FETCH-LOGICAL-c300t-33b857a9667940108fd1cb8fbcf39843d66a0db9c20eef8825e64da189674de53
ISICitedReferencesCount 10
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001229689200001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0957-4174
IngestDate Sat Nov 29 03:07:37 EST 2025
Tue Nov 18 22:12:49 EST 2025
Sat Aug 31 16:00:37 EDT 2024
IsPeerReviewed true
IsScholarly true
Keywords Causality
Relational learning
Incremental learning
Language English
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c300t-33b857a9667940108fd1cb8fbcf39843d66a0db9c20eef8825e64da189674de53
ORCID 0000-0002-0119-122X
0000-0003-2734-0622
0000-0003-3467-8963
0009-0008-4359-3383
0000-0002-4306-6945
0000-0001-9850-0270
0000-0002-7696-1444
ParticipantIDs crossref_primary_10_1016_j_eswa_2024_123901
crossref_citationtrail_10_1016_j_eswa_2024_123901
elsevier_sciencedirect_doi_10_1016_j_eswa_2024_123901
PublicationCentury 2000
PublicationDate 2024-09-15
PublicationDateYYYYMMDD 2024-09-15
PublicationDate_xml – month: 09
  year: 2024
  text: 2024-09-15
  day: 15
PublicationDecade 2020
PublicationTitle Expert systems with applications
PublicationYear 2024
Publisher Elsevier Ltd
Publisher_xml – name: Elsevier Ltd
References Pearl (b47) 2009; 3
Hu, Tang, Miao, Hua, Zhang (b22) 2021
Song, Lee, Kweon, Choi (b60) 2023
Maltoni, Lomonaco (b37) 2019; 116
Gong, Zhang, Liu, Tao, Glymour, Schölkopf (b16) 2016
Pearl (b46) 2009
Suárez-Cetrulo, Quintana, Cervantes (b62) 2023; 213
DeVries, Taylor (b14) 2017
Hernán, Robins (b19) 2010
Wang, Yi, Chen, Zhu (b71) 2022
Ven, Tolias (b66) 2019
Wang, Zhang, Yang, Yu, Li, Hong (b73) 2022
Hu, Gu, Zhang, Dai, Wei (b21) 2018
Mitrovic, McWilliams, Walker, Buesing, Blundell (b41) 2021
Bateman, Ngiam, Birney (b4) 2018; 13
Khosla, Teterwak, Wang, Sarna, Tian, Isola (b24) 2020
Zhao, Xiao, Gan, Zhang, Xia (b85) 2020
Chaudhry, Rohrbach, Elhoseiny, Ajanthan, Dokania, Torr (b12) 2019
Santoro, Raposo, Barrett, Malinowski, Pascanu, Battaglia (b56) 2017
Rebuffi, Kolesnikov, Sperl, Lampert (b54) 2017
Hinton, Vinyals, Dean (b20) 2015
Kang, Park, Han (b23) 2022
Xia, Yin, Feng, Yu (b77) 2021
Mai, Li, Kim, Sanner (b36) 2021
Krizhevsky, Hinton (b26) 2009
Waqas, Tahir, Qureshi (b74) 2023; 53
Mirzadeh, Farajtabar, Pascanu, Ghasemzadeh (b40) 2020; 33
Koch, Zemel, Salakhutdinov (b25) 2015
Student (b61) 1908
Xie, Li, He, Cao (b79) 2023; 225
Caccia, Aljundi, Asadi, Tuytelaars, Pineau, Belilovsky (b8) 2022
Benjamin, Rolnick, Kording (b5) 2019
Zhou, Wang, Qi, Ye, Zhan, Liu (b86) 2023
Chen, Kornblith, Norouzi, Hinton (b13) 2020
Zaadnoordijk, Besold, Cusack (b82) 2020
Aljundi, Belilovsky, Tuytelaars, Charlin, Caccia, Lin (b1) 2019
Yu, Twardowski, Liu, Herranz, Wang, Cheng (b81) 2020
Lin, Zhang, Feng, Li, Ye (b30) 2023
Xie, Dai, Hovy, Luong, Le (b78) 2020
Nguyen, Do, Nguyen, Duong, Nguyen (b42) 2023
Li, Peng, Li, Xia, Yang, Sun (b29) 2020
Wang, Yao, Kwok, Ni (b70) 2020; 53
Aljundi, Lin, Goujaud, Bengio (b2) 2019
McCloskey, Cohen (b39) 1989; vol. 24
Snell, Swersky, Zemel (b58) 2017
Tiwari, Killamsetty, Iyer, Shenoy (b65) 2022
Buzzega, Boschini, Porrello, Abati, Calderara (b6) 2020
Tang, Matteson (b64) 2021
Van der Maaten, Hinton (b33) 2008; 9
Chaudhry, Ranzato, Rohrbach, Elhoseiny (b11) 2019
Ratcliff (b53) 1990; 97 2
Mahajan, Tople, Sharma (b34) 2021
Qi, Brown, Lowe (b52) 2018
Prabhu, Torr, Dokania (b51) 2020
Dietterich (b15) 1998; 10
Sohn, Berthelot, Carlini, Zhang, Zhang, Raffel (b59) 2020
Wang, Zhang, Su, Zhu (b72) 2023
Rubin (b55) 2005; 100
Buzzega, Boschini, Porrello, Calderara (b7) 2020
Sung, Yang, Zhang, Xiang, Torr, Hospedales (b63) 2018
Zhang, Cisse, Dauphin, Lopez-Paz (b84) 2017
LeCun, Bengio, Hinton (b28) 2015; 521
Chaudhry, Gordo, Dokania, Torr, Lopez-Paz (b10) 2020
He, Zhang, Ren, Sun (b18) 2016
Krueger, Caballero, Jacobsen, Zhang, Binas, Zhang (b27) 2021
Pearl, Mackenzie (b48) 2018
Patacchiola, Storkey (b45) 2020
Gopalakrishnan, Singh, Fayek, Ramasamy, Ambikapathi (b17) 2022
Zenke, Poole, Ganguli (b83) 2017
Mai, Li, Jeong, Quispe, Kim, Sanner (b35) 2022; 469
Welling (b75) 2009
Peters, Janzing, Schlkopf (b50) 2017
Vinyals, Blundell, Lillicrap, Wierstra (b67) 2016; 29
Vitter (b68) 1985; 11
Arjovsky, Bottou, Gulrajani, Lopez-Paz (b3) 2019
Mao, Cha, Gupta, Wang, Yang, Vondrick (b38) 2021
Orhan, Gupta, Lake (b43) 2020; 33
Chaudhry, Dokania, Ajanthan, Torr (b9) 2018
Yan, Gong, Liu, van den Hengel, Shi (b80) 2022
Lopez-Paz, Ranzato (b32) 2017
Peters, Bühlmann, Meinshausen (b49) 2016
Zhou, Ye, Zhan (b87) 2021
Schölkopf (b57) 2019
Parisi, Kemker, Part, Kanan, Wermter (b44) 2019; 113
Llopis-Ibor, Beltran-Royo, Cuesta-Infante, Pantrigo (b31) 2023; 212
Von Kügelgen, Sharma, Gresele, Brendel, Schölkopf, Besserve (b69) 2021; 34
Wu, Chen, Wang, Ye, Liu, Guo (b76) 2019
Chaudhry (10.1016/j.eswa.2024.123901_b9) 2018
Koch (10.1016/j.eswa.2024.123901_b25) 2015
Mahajan (10.1016/j.eswa.2024.123901_b34) 2021
Tang (10.1016/j.eswa.2024.123901_b64) 2021
Li (10.1016/j.eswa.2024.123901_b29) 2020
Gong (10.1016/j.eswa.2024.123901_b16) 2016
Hu (10.1016/j.eswa.2024.123901_b22) 2021
Pearl (10.1016/j.eswa.2024.123901_b46) 2009
Schölkopf (10.1016/j.eswa.2024.123901_b57) 2019
Vinyals (10.1016/j.eswa.2024.123901_b67) 2016; 29
Yan (10.1016/j.eswa.2024.123901_b80) 2022
Xie (10.1016/j.eswa.2024.123901_b79) 2023; 225
Zhou (10.1016/j.eswa.2024.123901_b86) 2023
Mai (10.1016/j.eswa.2024.123901_b35) 2022; 469
Hu (10.1016/j.eswa.2024.123901_b21) 2018
Buzzega (10.1016/j.eswa.2024.123901_b6) 2020
Suárez-Cetrulo (10.1016/j.eswa.2024.123901_b62) 2023; 213
Zaadnoordijk (10.1016/j.eswa.2024.123901_b82) 2020
Ven (10.1016/j.eswa.2024.123901_b66) 2019
Chen (10.1016/j.eswa.2024.123901_b13) 2020
Dietterich (10.1016/j.eswa.2024.123901_b15) 1998; 10
LeCun (10.1016/j.eswa.2024.123901_b28) 2015; 521
Rubin (10.1016/j.eswa.2024.123901_b55) 2005; 100
Santoro (10.1016/j.eswa.2024.123901_b56) 2017
Zhao (10.1016/j.eswa.2024.123901_b85) 2020
Wang (10.1016/j.eswa.2024.123901_b71) 2022
Mao (10.1016/j.eswa.2024.123901_b38) 2021
Mirzadeh (10.1016/j.eswa.2024.123901_b40) 2020; 33
Waqas (10.1016/j.eswa.2024.123901_b74) 2023; 53
Caccia (10.1016/j.eswa.2024.123901_b8) 2022
Student (10.1016/j.eswa.2024.123901_b61) 1908
Von Kügelgen (10.1016/j.eswa.2024.123901_b69) 2021; 34
Orhan (10.1016/j.eswa.2024.123901_b43) 2020; 33
Zenke (10.1016/j.eswa.2024.123901_b83) 2017
Arjovsky (10.1016/j.eswa.2024.123901_b3) 2019
Patacchiola (10.1016/j.eswa.2024.123901_b45) 2020
Bateman (10.1016/j.eswa.2024.123901_b4) 2018; 13
Wu (10.1016/j.eswa.2024.123901_b76) 2019
Chaudhry (10.1016/j.eswa.2024.123901_b12) 2019
DeVries (10.1016/j.eswa.2024.123901_b14) 2017
Yu (10.1016/j.eswa.2024.123901_b81) 2020
Benjamin (10.1016/j.eswa.2024.123901_b5) 2019
Tiwari (10.1016/j.eswa.2024.123901_b65) 2022
Maltoni (10.1016/j.eswa.2024.123901_b37) 2019; 116
Wang (10.1016/j.eswa.2024.123901_b72) 2023
Rebuffi (10.1016/j.eswa.2024.123901_b54) 2017
Vitter (10.1016/j.eswa.2024.123901_b68) 1985; 11
Prabhu (10.1016/j.eswa.2024.123901_b51) 2020
McCloskey (10.1016/j.eswa.2024.123901_b39) 1989; vol. 24
Zhang (10.1016/j.eswa.2024.123901_b84) 2017
Llopis-Ibor (10.1016/j.eswa.2024.123901_b31) 2023; 212
Ratcliff (10.1016/j.eswa.2024.123901_b53) 1990; 97 2
Krueger (10.1016/j.eswa.2024.123901_b27) 2021
Lin (10.1016/j.eswa.2024.123901_b30) 2023
Pearl (10.1016/j.eswa.2024.123901_b48) 2018
Snell (10.1016/j.eswa.2024.123901_b58) 2017
Welling (10.1016/j.eswa.2024.123901_b75) 2009
Khosla (10.1016/j.eswa.2024.123901_b24) 2020
Pearl (10.1016/j.eswa.2024.123901_b47) 2009; 3
Wang (10.1016/j.eswa.2024.123901_b73) 2022
Gopalakrishnan (10.1016/j.eswa.2024.123901_b17) 2022
Aljundi (10.1016/j.eswa.2024.123901_b1) 2019
Lopez-Paz (10.1016/j.eswa.2024.123901_b32) 2017
Zhou (10.1016/j.eswa.2024.123901_b87) 2021
Krizhevsky (10.1016/j.eswa.2024.123901_b26) 2009
Mitrovic (10.1016/j.eswa.2024.123901_b41) 2021
Xie (10.1016/j.eswa.2024.123901_b78) 2020
Hinton (10.1016/j.eswa.2024.123901_b20) 2015
Sung (10.1016/j.eswa.2024.123901_b63) 2018
Mai (10.1016/j.eswa.2024.123901_b36) 2021
Parisi (10.1016/j.eswa.2024.123901_b44) 2019; 113
Xia (10.1016/j.eswa.2024.123901_b77) 2021
Peters (10.1016/j.eswa.2024.123901_b49) 2016
Peters (10.1016/j.eswa.2024.123901_b50) 2017
Hernán (10.1016/j.eswa.2024.123901_b19) 2010
Van der Maaten (10.1016/j.eswa.2024.123901_b33) 2008; 9
Chaudhry (10.1016/j.eswa.2024.123901_b11) 2019
Qi (10.1016/j.eswa.2024.123901_b52) 2018
Buzzega (10.1016/j.eswa.2024.123901_b7) 2020
He (10.1016/j.eswa.2024.123901_b18) 2016
Sohn (10.1016/j.eswa.2024.123901_b59) 2020
Chaudhry (10.1016/j.eswa.2024.123901_b10) 2020
Song (10.1016/j.eswa.2024.123901_b60) 2023
Kang (10.1016/j.eswa.2024.123901_b23) 2022
Nguyen (10.1016/j.eswa.2024.123901_b42) 2023
Wang (10.1016/j.eswa.2024.123901_b70) 2020; 53
Aljundi (10.1016/j.eswa.2024.123901_b2) 2019
References_xml – year: 2021
  ident: b64
  article-title: Graph-based continual learning
  publication-title: ICLR
– volume: 34
  start-page: 16451
  year: 2021
  end-page: 16467
  ident: b69
  article-title: Self-supervised learning with data augmentations provably isolates content from style
  publication-title: NeurIPS
– year: 2017
  ident: b14
  article-title: Improved regularization of convolutional neural networks with cutout
– year: 2010
  ident: b19
  article-title: Causal inference
– year: 2023
  ident: b86
  article-title: Deep class-incremental learning: A survey
– start-page: 6256
  year: 2020
  end-page: 6268
  ident: b78
  article-title: Unsupervised data augmentation for consistency training
  publication-title: NeurIPS
– volume: 33
  start-page: 9960
  year: 2020
  end-page: 9971
  ident: b43
  article-title: Self-supervised learning through the eyes of a child
  publication-title: NeurIPS
– volume: 521
  start-page: 436
  year: 2015
  end-page: 444
  ident: b28
  article-title: Deep learning
  publication-title: Nature
– volume: 116
  start-page: 56
  year: 2019
  end-page: 73
  ident: b37
  article-title: Continuous learning in single-incremental-task scenarios
  publication-title: Neural Networks
– start-page: 1199
  year: 2018
  end-page: 1208
  ident: b63
  article-title: Learning to compare: Relation network for few-shot learning
  publication-title: CVPR
– start-page: 374
  year: 2019
  end-page: 382
  ident: b76
  article-title: Large scale incremental learning
  publication-title: CVPR
– year: 2020
  ident: b29
  article-title: A survey on text classification: From shallow to deep learning
– volume: vol. 24
  start-page: 109
  year: 1989
  end-page: 165
  ident: b39
  publication-title: Catastrophic interference in connectionist networks: the sequential learning problem
– volume: 13
  year: 2018
  ident: b4
  article-title: Relational encoding of objects in working memory: Changes detection performance is better for violations in group relations
  publication-title: PLoS One
– volume: 9
  year: 2008
  ident: b33
  article-title: Visualizing data using t-SNE
  publication-title: JMLR
– year: 2017
  ident: b58
  article-title: Prototypical networks for few-shot learning
  publication-title: NIPS
– year: 2020
  ident: b10
  article-title: Using hindsight to anchor past knowledge in continual learning
– volume: 29
  year: 2016
  ident: b67
  article-title: Matching networks for one shot learning
  publication-title: NIPS
– start-page: 2180
  year: 2020
  end-page: 2187
  ident: b7
  article-title: Rethinking experience replay: a bag of tricks for continual learning
  publication-title: ICPR
– year: 2021
  ident: b41
  article-title: Representation learning via invariant causal mechanisms
  publication-title: ICLR
– start-page: 375
  year: 2022
  end-page: 385
  ident: b71
  article-title: Out-of-distribution generalization with causal invariant transformations
  publication-title: CVPR
– volume: 469
  start-page: 28
  year: 2022
  end-page: 51
  ident: b35
  article-title: Online continual learning in image classification: An empirical survey
  publication-title: Neurocomputing
– start-page: 6470
  year: 2017
  end-page: 6479
  ident: b32
  article-title: Gradient episodic memory for continual learning
  publication-title: NIPS
– start-page: 16071
  year: 2022
  end-page: 16080
  ident: b23
  article-title: Class-incremental learning by knowledge distillation with adaptive feature consolidation
  publication-title: CVPR
– year: 2017
  ident: b50
  article-title: Elements of causal inference: foundations and learning algorithms
– year: 2018
  ident: b48
  article-title: The book of why: the new science of cause and effect
– volume: 97 2
  start-page: 285
  year: 1990
  end-page: 308
  ident: b53
  article-title: Connectionist models of recognition memory: constraints imposed by learning and forgetting functions
  publication-title: Psychological Review
– year: 2019
  ident: b66
  article-title: Three scenarios for continual learning
– volume: 53
  start-page: 10310
  year: 2023
  end-page: 10325
  ident: b74
  article-title: Deep Gaussian mixture model based instance relevance estimation for multiple instance learning applications
  publication-title: Applied Intelligence
– start-page: 532
  year: 2018
  end-page: 547
  ident: b9
  article-title: Riemannian walk for incremental learning: Understanding forgetting and intransigence
  publication-title: ECCV
– volume: 53
  start-page: 1
  year: 2020
  end-page: 34
  ident: b70
  article-title: Generalizing from a few examples: A survey on few-shot learning
  publication-title: ACM Computing Surveys
– start-page: 1597
  year: 2020
  end-page: 1607
  ident: b13
  article-title: A simple framework for contrastive learning of visual representations
  publication-title: ICML
– start-page: 1
  year: 2020
  end-page: 11
  ident: b6
  article-title: Dark experience for general continual learning: a strong, simple baseline
  publication-title: NeurIPS
– volume: 213
  year: 2023
  ident: b62
  article-title: A survey on machine learning for recurring concept drifting data streams
  publication-title: Expert Systems with Applications
– start-page: 4967
  year: 2017
  end-page: 4976
  ident: b56
  article-title: A simple neural network module for relational reasoning
  publication-title: NIPS
– year: 2020
  ident: b82
  article-title: The next big thing (s) in unsupervised machine learning: Five lessons from infant learning
– start-page: 1
  year: 2020
  end-page: 13
  ident: b24
  article-title: Supervised contrastive learning
  publication-title: NeurIPS
– year: 2023
  ident: b72
  article-title: A comprehensive survey of continual learning: Theory, method and application
– start-page: 947
  year: 2016
  end-page: 1012
  ident: b49
  article-title: Causal inference by using invariant prediction: identification and confidence intervals
  publication-title: Journal of the Royal Statistical Society. Series B (Statistical Methodology)
– start-page: 524
  year: 2020
  end-page: 540
  ident: b51
  article-title: Gdumb: A simple approach that questions our progress in continual learning
  publication-title: ECCV
– year: 2009
  ident: b46
  article-title: Causality: models, reasoning and inference
– year: 2019
  ident: b3
  article-title: Invariant risk minimization
– year: 2015
  ident: b20
  article-title: Distilling the knowledge in a neural network
– year: 2022
  ident: b73
  article-title: Memory replay with data compression for continual learning
  publication-title: ICLR
– start-page: 24246
  year: 2023
  end-page: 24255
  ident: b30
  article-title: PCR: Proxy-based contrastive replay for online class-incremental continual learning
  publication-title: CVPR
– start-page: 5822
  year: 2018
  end-page: 5830
  ident: b52
  article-title: Low-shot learning with imprinted weights
  publication-title: CVPR
– start-page: 3987
  year: 2017
  end-page: 3995
  ident: b83
  article-title: Continual learning through synaptic intelligence
  publication-title: ICML
– start-page: 3588
  year: 2018
  end-page: 3597
  ident: b21
  article-title: Relation networks for object detection
  publication-title: CVPR
– volume: 225
  year: 2023
  ident: b79
  article-title: Few-shot class-incremental audio classification via discriminative prototype learning
  publication-title: Expert Systems with Applications
– start-page: 11849
  year: 2019
  end-page: 11860
  ident: b1
  article-title: Online continual learning with maximal interfered retrieval
  publication-title: NeurIPS
– year: 2022
  ident: b8
  article-title: New insights on reducing abrupt representation change in online continual learning
  publication-title: ICLR
– start-page: 3584
  year: 2021
  end-page: 3594
  ident: b36
  article-title: Supervised contrastive replay: Revisiting the nearest class mean classifier in online class-incremental continual learning
  publication-title: CVPRW
– start-page: 2839
  year: 2016
  end-page: 2848
  ident: b16
  article-title: Domain adaptation with conditional transferable components
  publication-title: ICML
– volume: 113
  start-page: 54
  year: 2019
  end-page: 71
  ident: b44
  article-title: Continual lifelong learning with neural networks: A review
  publication-title: Neural Networks
– start-page: 7313
  year: 2021
  end-page: 7324
  ident: b34
  article-title: Domain generalization using causal matching
  publication-title: ICML
– year: 2017
  ident: b84
  article-title: Mixup: Beyond empirical risk minimization
– volume: 3
  start-page: 96
  year: 2009
  end-page: 146
  ident: b47
  article-title: Causal inference in statistics: An overview
  publication-title: Statistics Surveys
– start-page: 1645
  year: 2021
  end-page: 1654
  ident: b87
  article-title: Co-transport for class-incremental learning
  publication-title: ACMMM
– start-page: 10
  year: 2022
  end-page: 18
  ident: b17
  article-title: Knowledge capture and replay for continual learning
  publication-title: WACV
– start-page: 1746
  year: 2023
  end-page: 1757
  ident: b42
  article-title: Causal inference via style transfer for out-of-distribution generalisation
  publication-title: KDD
– start-page: 3947
  year: 2021
  end-page: 3956
  ident: b38
  article-title: Generative interventions for causal learning
  publication-title: CVPR
– start-page: 596
  year: 2020
  end-page: 608
  ident: b59
  article-title: FixMatch: Simplifying semi-supervised learning with consistency and confidence
  publication-title: NeurIPS
– year: 2009
  ident: b26
  article-title: Learning multiple layers of features from tiny images
– start-page: 4003
  year: 2020
  end-page: 4014
  ident: b45
  article-title: Self-supervised relational reasoning for representation learning
  publication-title: NeurIPS
– volume: 11
  start-page: 37
  year: 1985
  end-page: 57
  ident: b68
  article-title: Random sampling with a reservoir
  publication-title: ACM Transactions on Mathematical Software
– start-page: 770
  year: 2016
  end-page: 778
  ident: b18
  article-title: Deep residual learning for image recognition
  publication-title: CVPR
– start-page: 109
  year: 2022
  end-page: 118
  ident: b80
  article-title: Learning Bayesian sparse networks with full experience replay for continual learning
  publication-title: CVPR
– start-page: 6982
  year: 2020
  end-page: 6991
  ident: b81
  article-title: Semantic drift compensation for class-incremental learning
  publication-title: CVPR
– year: 2019
  ident: b5
  article-title: Measuring and regularizing networks in function space
  publication-title: ICLR
– start-page: 1
  year: 2019
  end-page: 20
  ident: b11
  article-title: Efficient lifelong learning with A-GEM
  publication-title: ICLR
– volume: 212
  year: 2023
  ident: b31
  article-title: Fast incremental learning by transfer learning and hierarchical sequencing
  publication-title: Expert Systems with Applications
– start-page: 99
  year: 2022
  end-page: 108
  ident: b65
  article-title: Gcr: Gradient coreset based replay buffer selection for continual learning
  publication-title: CVPR
– start-page: 1121
  year: 2009
  end-page: 1128
  ident: b75
  article-title: Herding dynamical weights to learn
  publication-title: ICML
– year: 2015
  ident: b25
  article-title: Siamese neural networks for one-shot image recognition
  publication-title: ICML workshop
– start-page: 5815
  year: 2021
  end-page: 5826
  ident: b27
  article-title: Out-of-distribution generalization via risk extrapolation (rex)
  publication-title: ICML
– start-page: 13208
  year: 2020
  end-page: 13217
  ident: b85
  article-title: Maintaining discrimination and fairness in class incremental learning
  publication-title: CVPR
– start-page: 11920
  year: 2023
  end-page: 11929
  ident: b60
  article-title: EcoTTA: Memory-efficient continual test-time adaptation via self-distilled regularization
  publication-title: CVPR
– volume: 100
  start-page: 322
  year: 2005
  end-page: 331
  ident: b55
  article-title: Causal inference using potential outcomes: Design, modeling, decisions
  publication-title: Journal of the American Statistical Association
– volume: 33
  start-page: 7308
  year: 2020
  end-page: 7320
  ident: b40
  article-title: Understanding the role of training regimes in continual learning
  publication-title: NeurIPS
– start-page: 11816
  year: 2019
  end-page: 11825
  ident: b2
  article-title: Gradient based sample selection for online continual learning
  publication-title: NeurIPS
– start-page: 1
  year: 1908
  end-page: 25
  ident: b61
  article-title: The probable error of a mean
  publication-title: Biometrika
– year: 2021
  ident: b77
  article-title: Incremental few-shot text classification with multi-round new classes: Formulation, dataset and system
– start-page: 2001
  year: 2017
  end-page: 2010
  ident: b54
  article-title: ICaRL: Incremental classifier and representation learning
  publication-title: CVPR
– year: 2019
  ident: b57
  article-title: Causality for machine learning
– start-page: 3957
  year: 2021
  end-page: 3966
  ident: b22
  article-title: Distilling causal effect of data in class-incremental learning
  publication-title: CVPR
– year: 2019
  ident: b12
  article-title: On tiny episodic memories in continual learning
– volume: 10
  start-page: 1895
  year: 1998
  end-page: 1923
  ident: b15
  article-title: Approximate statistical tests for comparing supervised classification learning algorithms
  publication-title: Neural Computation
– volume: 97 2
  start-page: 285
  year: 1990
  ident: 10.1016/j.eswa.2024.123901_b53
  article-title: Connectionist models of recognition memory: constraints imposed by learning and forgetting functions
  publication-title: Psychological Review
  doi: 10.1037/0033-295X.97.2.285
– year: 2022
  ident: 10.1016/j.eswa.2024.123901_b73
  article-title: Memory replay with data compression for continual learning
– year: 2018
  ident: 10.1016/j.eswa.2024.123901_b48
– start-page: 3957
  year: 2021
  ident: 10.1016/j.eswa.2024.123901_b22
  article-title: Distilling causal effect of data in class-incremental learning
– start-page: 1
  year: 1908
  ident: 10.1016/j.eswa.2024.123901_b61
  article-title: The probable error of a mean
  publication-title: Biometrika
  doi: 10.2307/2331554
– start-page: 16071
  year: 2022
  ident: 10.1016/j.eswa.2024.123901_b23
  article-title: Class-incremental learning by knowledge distillation with adaptive feature consolidation
– year: 2021
  ident: 10.1016/j.eswa.2024.123901_b64
  article-title: Graph-based continual learning
– start-page: 770
  year: 2016
  ident: 10.1016/j.eswa.2024.123901_b18
  article-title: Deep residual learning for image recognition
– start-page: 11849
  year: 2019
  ident: 10.1016/j.eswa.2024.123901_b1
  article-title: Online continual learning with maximal interfered retrieval
– start-page: 947
  year: 2016
  ident: 10.1016/j.eswa.2024.123901_b49
  article-title: Causal inference by using invariant prediction: identification and confidence intervals
  publication-title: Journal of the Royal Statistical Society. Series B (Statistical Methodology)
  doi: 10.1111/rssb.12167
– start-page: 3588
  year: 2018
  ident: 10.1016/j.eswa.2024.123901_b21
  article-title: Relation networks for object detection
– start-page: 5815
  year: 2021
  ident: 10.1016/j.eswa.2024.123901_b27
  article-title: Out-of-distribution generalization via risk extrapolation (rex)
– volume: 33
  start-page: 9960
  year: 2020
  ident: 10.1016/j.eswa.2024.123901_b43
  article-title: Self-supervised learning through the eyes of a child
  publication-title: NeurIPS
– volume: 225
  year: 2023
  ident: 10.1016/j.eswa.2024.123901_b79
  article-title: Few-shot class-incremental audio classification via discriminative prototype learning
  publication-title: Expert Systems with Applications
  doi: 10.1016/j.eswa.2023.120044
– start-page: 1
  year: 2020
  ident: 10.1016/j.eswa.2024.123901_b6
  article-title: Dark experience for general continual learning: a strong, simple baseline
– year: 2021
  ident: 10.1016/j.eswa.2024.123901_b41
  article-title: Representation learning via invariant causal mechanisms
– start-page: 524
  year: 2020
  ident: 10.1016/j.eswa.2024.123901_b51
  article-title: Gdumb: A simple approach that questions our progress in continual learning
– start-page: 99
  year: 2022
  ident: 10.1016/j.eswa.2024.123901_b65
  article-title: Gcr: Gradient coreset based replay buffer selection for continual learning
– volume: 3
  start-page: 96
  year: 2009
  ident: 10.1016/j.eswa.2024.123901_b47
  article-title: Causal inference in statistics: An overview
  publication-title: Statistics Surveys
  doi: 10.1214/09-SS057
– start-page: 1121
  year: 2009
  ident: 10.1016/j.eswa.2024.123901_b75
  article-title: Herding dynamical weights to learn
– volume: 11
  start-page: 37
  issue: 1
  year: 1985
  ident: 10.1016/j.eswa.2024.123901_b68
  article-title: Random sampling with a reservoir
  publication-title: ACM Transactions on Mathematical Software
  doi: 10.1145/3147.3165
– volume: 33
  start-page: 7308
  year: 2020
  ident: 10.1016/j.eswa.2024.123901_b40
  article-title: Understanding the role of training regimes in continual learning
  publication-title: NeurIPS
– start-page: 596
  year: 2020
  ident: 10.1016/j.eswa.2024.123901_b59
  article-title: FixMatch: Simplifying semi-supervised learning with consistency and confidence
– year: 2017
  ident: 10.1016/j.eswa.2024.123901_b14
– volume: 53
  start-page: 10310
  issue: 9
  year: 2023
  ident: 10.1016/j.eswa.2024.123901_b74
  article-title: Deep Gaussian mixture model based instance relevance estimation for multiple instance learning applications
  publication-title: Applied Intelligence
  doi: 10.1007/s10489-022-04045-7
– year: 2020
  ident: 10.1016/j.eswa.2024.123901_b10
– start-page: 1
  year: 2020
  ident: 10.1016/j.eswa.2024.123901_b24
  article-title: Supervised contrastive learning
– year: 2020
  ident: 10.1016/j.eswa.2024.123901_b29
– start-page: 1
  year: 2019
  ident: 10.1016/j.eswa.2024.123901_b11
  article-title: Efficient lifelong learning with A-GEM
– start-page: 374
  year: 2019
  ident: 10.1016/j.eswa.2024.123901_b76
  article-title: Large scale incremental learning
– year: 2015
  ident: 10.1016/j.eswa.2024.123901_b20
– year: 2019
  ident: 10.1016/j.eswa.2024.123901_b5
  article-title: Measuring and regularizing networks in function space
– volume: 10
  start-page: 1895
  issue: 7
  year: 1998
  ident: 10.1016/j.eswa.2024.123901_b15
  article-title: Approximate statistical tests for comparing supervised classification learning algorithms
  publication-title: Neural Computation
  doi: 10.1162/089976698300017197
– start-page: 6256
  year: 2020
  ident: 10.1016/j.eswa.2024.123901_b78
  article-title: Unsupervised data augmentation for consistency training
– start-page: 532
  year: 2018
  ident: 10.1016/j.eswa.2024.123901_b9
  article-title: Riemannian walk for incremental learning: Understanding forgetting and intransigence
– year: 2019
  ident: 10.1016/j.eswa.2024.123901_b3
– volume: 34
  start-page: 16451
  year: 2021
  ident: 10.1016/j.eswa.2024.123901_b69
  article-title: Self-supervised learning with data augmentations provably isolates content from style
  publication-title: NeurIPS
– start-page: 13208
  year: 2020
  ident: 10.1016/j.eswa.2024.123901_b85
  article-title: Maintaining discrimination and fairness in class incremental learning
– volume: 9
  issue: 11
  year: 2008
  ident: 10.1016/j.eswa.2024.123901_b33
  article-title: Visualizing data using t-SNE
  publication-title: JMLR
– year: 2023
  ident: 10.1016/j.eswa.2024.123901_b72
– year: 2021
  ident: 10.1016/j.eswa.2024.123901_b77
– volume: 521
  start-page: 436
  issue: 7553
  year: 2015
  ident: 10.1016/j.eswa.2024.123901_b28
  article-title: Deep learning
  publication-title: Nature
  doi: 10.1038/nature14539
– year: 2017
  ident: 10.1016/j.eswa.2024.123901_b50
– year: 2020
  ident: 10.1016/j.eswa.2024.123901_b82
– start-page: 1645
  year: 2021
  ident: 10.1016/j.eswa.2024.123901_b87
  article-title: Co-transport for class-incremental learning
– start-page: 4967
  year: 2017
  ident: 10.1016/j.eswa.2024.123901_b56
  article-title: A simple neural network module for relational reasoning
– start-page: 5822
  year: 2018
  ident: 10.1016/j.eswa.2024.123901_b52
  article-title: Low-shot learning with imprinted weights
– volume: 100
  start-page: 322
  issue: 469
  year: 2005
  ident: 10.1016/j.eswa.2024.123901_b55
  article-title: Causal inference using potential outcomes: Design, modeling, decisions
  publication-title: Journal of the American Statistical Association
  doi: 10.1198/016214504000001880
– start-page: 7313
  year: 2021
  ident: 10.1016/j.eswa.2024.123901_b34
  article-title: Domain generalization using causal matching
– start-page: 11920
  year: 2023
  ident: 10.1016/j.eswa.2024.123901_b60
  article-title: EcoTTA: Memory-efficient continual test-time adaptation via self-distilled regularization
– start-page: 2180
  year: 2020
  ident: 10.1016/j.eswa.2024.123901_b7
  article-title: Rethinking experience replay: a bag of tricks for continual learning
– year: 2019
  ident: 10.1016/j.eswa.2024.123901_b12
– start-page: 10
  year: 2022
  ident: 10.1016/j.eswa.2024.123901_b17
  article-title: Knowledge capture and replay for continual learning
– year: 2019
  ident: 10.1016/j.eswa.2024.123901_b57
– start-page: 3987
  year: 2017
  ident: 10.1016/j.eswa.2024.123901_b83
  article-title: Continual learning through synaptic intelligence
– start-page: 2001
  year: 2017
  ident: 10.1016/j.eswa.2024.123901_b54
  article-title: ICaRL: Incremental classifier and representation learning
– year: 2017
  ident: 10.1016/j.eswa.2024.123901_b58
  article-title: Prototypical networks for few-shot learning
– volume: 212
  year: 2023
  ident: 10.1016/j.eswa.2024.123901_b31
  article-title: Fast incremental learning by transfer learning and hierarchical sequencing
  publication-title: Expert Systems with Applications
  doi: 10.1016/j.eswa.2022.118580
– start-page: 3584
  year: 2021
  ident: 10.1016/j.eswa.2024.123901_b36
  article-title: Supervised contrastive replay: Revisiting the nearest class mean classifier in online class-incremental continual learning
– start-page: 1746
  year: 2023
  ident: 10.1016/j.eswa.2024.123901_b42
  article-title: Causal inference via style transfer for out-of-distribution generalisation
– start-page: 2839
  year: 2016
  ident: 10.1016/j.eswa.2024.123901_b16
  article-title: Domain adaptation with conditional transferable components
– volume: 53
  start-page: 1
  issue: 3
  year: 2020
  ident: 10.1016/j.eswa.2024.123901_b70
  article-title: Generalizing from a few examples: A survey on few-shot learning
  publication-title: ACM Computing Surveys
  doi: 10.1145/3386252
– volume: 116
  start-page: 56
  year: 2019
  ident: 10.1016/j.eswa.2024.123901_b37
  article-title: Continuous learning in single-incremental-task scenarios
  publication-title: Neural Networks
  doi: 10.1016/j.neunet.2019.03.010
– volume: 213
  year: 2023
  ident: 10.1016/j.eswa.2024.123901_b62
  article-title: A survey on machine learning for recurring concept drifting data streams
  publication-title: Expert Systems with Applications
  doi: 10.1016/j.eswa.2022.118934
– start-page: 4003
  year: 2020
  ident: 10.1016/j.eswa.2024.123901_b45
  article-title: Self-supervised relational reasoning for representation learning
– start-page: 11816
  year: 2019
  ident: 10.1016/j.eswa.2024.123901_b2
  article-title: Gradient based sample selection for online continual learning
– start-page: 24246
  year: 2023
  ident: 10.1016/j.eswa.2024.123901_b30
  article-title: PCR: Proxy-based contrastive replay for online class-incremental continual learning
– start-page: 6470
  year: 2017
  ident: 10.1016/j.eswa.2024.123901_b32
  article-title: Gradient episodic memory for continual learning
– volume: 469
  start-page: 28
  year: 2022
  ident: 10.1016/j.eswa.2024.123901_b35
  article-title: Online continual learning in image classification: An empirical survey
  publication-title: Neurocomputing
  doi: 10.1016/j.neucom.2021.10.021
– start-page: 375
  year: 2022
  ident: 10.1016/j.eswa.2024.123901_b71
  article-title: Out-of-distribution generalization with causal invariant transformations
– year: 2017
  ident: 10.1016/j.eswa.2024.123901_b84
– volume: 29
  year: 2016
  ident: 10.1016/j.eswa.2024.123901_b67
  article-title: Matching networks for one shot learning
  publication-title: NIPS
– volume: 13
  issue: 9
  year: 2018
  ident: 10.1016/j.eswa.2024.123901_b4
  article-title: Relational encoding of objects in working memory: Changes detection performance is better for violations in group relations
  publication-title: PLoS One
  doi: 10.1371/journal.pone.0203848
– year: 2010
  ident: 10.1016/j.eswa.2024.123901_b19
– start-page: 109
  year: 2022
  ident: 10.1016/j.eswa.2024.123901_b80
  article-title: Learning Bayesian sparse networks with full experience replay for continual learning
– start-page: 6982
  year: 2020
  ident: 10.1016/j.eswa.2024.123901_b81
  article-title: Semantic drift compensation for class-incremental learning
– year: 2019
  ident: 10.1016/j.eswa.2024.123901_b66
– start-page: 1597
  year: 2020
  ident: 10.1016/j.eswa.2024.123901_b13
  article-title: A simple framework for contrastive learning of visual representations
– year: 2015
  ident: 10.1016/j.eswa.2024.123901_b25
  article-title: Siamese neural networks for one-shot image recognition
– year: 2009
  ident: 10.1016/j.eswa.2024.123901_b46
– year: 2022
  ident: 10.1016/j.eswa.2024.123901_b8
  article-title: New insights on reducing abrupt representation change in online continual learning
– start-page: 3947
  year: 2021
  ident: 10.1016/j.eswa.2024.123901_b38
  article-title: Generative interventions for causal learning
– volume: 113
  start-page: 54
  year: 2019
  ident: 10.1016/j.eswa.2024.123901_b44
  article-title: Continual lifelong learning with neural networks: A review
  publication-title: Neural Networks
  doi: 10.1016/j.neunet.2019.01.012
– volume: vol. 24
  start-page: 109
  year: 1989
  ident: 10.1016/j.eswa.2024.123901_b39
– year: 2023
  ident: 10.1016/j.eswa.2024.123901_b86
– year: 2009
  ident: 10.1016/j.eswa.2024.123901_b26
– start-page: 1199
  year: 2018
  ident: 10.1016/j.eswa.2024.123901_b63
  article-title: Learning to compare: Relation network for few-shot learning
SSID ssj0017007
Score 2.4949348
Snippet In Class-Incremental Learning (Class-IL), deep neural networks often fail to learn a sequence of classes incrementally due to catastrophic forgetting, a...
SourceID crossref
elsevier
SourceType Enrichment Source
Index Database
Publisher
StartPage 123901
SubjectTerms Causality
Incremental learning
Relational learning
Title Class-incremental learning with causal relational replay
URI https://dx.doi.org/10.1016/j.eswa.2024.123901
Volume 250
WOSCitedRecordID wos001229689200001&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
  issn: 0957-4174
  databaseCode: AIEXJ
  dateStart: 19950101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://www.sciencedirect.com
  omitProxy: false
  ssIdentifier: ssj0017007
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1LT9wwELYo9NALr7YCCiiH3iqjPJzYPlIEAoQQUrfV3iLHsSvQKrva3VD67xln7N3lIdQicYmiyHai-azJzOd5EPKVcWUzJXJaSyso0zynMuea2jRRPNOFTTse8tcFv7wU_b688p0KJ107Ad404u5Ojt4UangGYLvU2f-Ae7YoPIB7AB2uADtc_wn4rs0lvW40Mn8AwSDQHxhnrtpJV81_EHjAsRkN1IPj3a4A8tSXeQ4JcAtH3TMK-Xf7F_VWb6gWzvNN2-myVs_tZB_6-10NHw_rtf7v6cmHlLlICUy_REYsZMXMQ5CQWuSUJdh9J2jZFOvLPtHYSB7cHJjJH1cGKmUH8C-VnuB4WAn7h1vYreviXnnB--_ISspzCcps5fDsuH8-Oz7iMebJhw_x2VIY2Pf4Tc9bJAtWRm-drHr3IDpEWDfIkmk2yVpovRF5TfyRiCcoRwHlyKEVIcrRHOUIUf5Efp4c945OqW-CQXUWx1OaZZXIuQKvFDQnOM_C1omuhK20zaRgWV0UKq4rqdPYGAv-Um4KVqtEyIKz2uTZZ7LcDBuzRSJVSFWD-WFSWzFVxVJx6ww0lmgFXibfJkkQRKl9hXjXqGRQhlDAm9IJr3TCK1F42-TbbM4I66O8ODoP8i29hYeWWwnb4YV5O6-c94V8mO_aXbI8Hbdmj7zXt9PryXjf75p7f3J1mw
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=Class-incremental+learning+with+causal+relational+replay&rft.jtitle=Expert+systems+with+applications&rft.au=Nguyen%2C+Toan&rft.au=Kieu%2C+Duc&rft.au=Duong%2C+Bao&rft.au=Kieu%2C+Tung&rft.date=2024-09-15&rft.pub=Elsevier+Ltd&rft.issn=0957-4174&rft.volume=250&rft_id=info:doi/10.1016%2Fj.eswa.2024.123901&rft.externalDocID=S095741742400767X
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0957-4174&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0957-4174&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0957-4174&client=summon