BEERL: Both Ends Explanations for Reinforcement Learning

Deep Reinforcement Learning (RL) is a black-box method and is hard to understand because the agent employs a neural network (NN). To explain the behavior and decisions made by the agent, different eXplainable RL (XRL) methods are developed; for example, feature importance methods are applied to anal...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Applied sciences Jg. 12; H. 21; S. 10947
Hauptverfasser: Terra, Ahmad, Inam, Rafia, Fersman, Elena
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Basel MDPI AG 01.11.2022
Schlagworte:
ISSN:2076-3417, 2076-3417
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract Deep Reinforcement Learning (RL) is a black-box method and is hard to understand because the agent employs a neural network (NN). To explain the behavior and decisions made by the agent, different eXplainable RL (XRL) methods are developed; for example, feature importance methods are applied to analyze the contribution of the input side of the model, and reward decomposition methods are applied to explain the components of the output end of the RL model. In this study, we present a novel method to connect explanations from both input and output ends of a black-box model, which results in fine-grained explanations. Our method exposes the reward prioritization to the user, which in turn generates two different levels of explanation and allows RL agent reconfigurations when unwanted behaviors are observed. The method further summarizes the detailed explanations into a focus value that takes into account all reward components and quantifies the fulfillment of the explanation of desired properties. We evaluated our method by applying it to a remote electrical telecom-antenna-tilt use case and two openAI gym environments: lunar lander and cartpole. The results demonstrated fine-grained explanations by detailing input features’ contributions to certain rewards and revealed biases of the reward components, which are then addressed by adjusting the reward’s weights.
AbstractList Deep Reinforcement Learning (RL) is a black-box method and is hard to understand because the agent employs a neural network (NN). To explain the behavior and decisions made by the agent, different eXplainable RL (XRL) methods are developed; for example, feature importance methods are applied to analyze the contribution of the input side of the model, and reward decomposition methods are applied to explain the components of the output end of the RL model. In this study, we present a novel method to connect explanations from both input and output ends of a black-box model, which results in fine-grained explanations. Our method exposes the reward prioritization to the user, which in turn generates two different levels of explanation and allows RL agent reconfigurations when unwanted behaviors are observed. The method further summarizes the detailed explanations into a focus value that takes into account all reward components and quantifies the fulfillment of the explanation of desired properties. We evaluated our method by applying it to a remote electrical telecom-antenna-tilt use case and two openAI gym environments: lunar lander and cartpole. The results demonstrated fine-grained explanations by detailing input features’ contributions to certain rewards and revealed biases of the reward components, which are then addressed by adjusting the reward’s weights.
Author Inam, Rafia
Fersman, Elena
Terra, Ahmad
Author_xml – sequence: 1
  givenname: Ahmad
  orcidid: 0000-0002-6650-2789
  surname: Terra
  fullname: Terra, Ahmad
– sequence: 2
  givenname: Rafia
  orcidid: 0000-0001-7448-3381
  surname: Inam
  fullname: Inam, Rafia
– sequence: 3
  givenname: Elena
  orcidid: 0000-0002-0182-8390
  surname: Fersman
  fullname: Fersman, Elena
BackLink https://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-329022$$DView record from Swedish Publication Index (Kungliga Tekniska Högskolan)
BookMark eNptkVtrVDEURoNUsLZ98wcc8NWjuV986-VUCwOFor6GXNuM0-SYZFD_fY8zCq00L18Iay_2zn4NDnLJAYA3CL4nRMEPZp4RxghBRcULcIih4COhSBw8ur8CJ62t4XIUIhLBQyDPpulm9XE4K_1umLJvw_Rr3phseiq5DbHU4SakvKQL9yH3YRVMzSnfHoOX0WxaOPmbR-Dr5fTl_PO4uv50dX66Gh2FrI9USieZ4RS7qLBl3HMrOXMRBRKVt5EEToXE3lOOWbDSSBFl9EogEiz35Ahc7b2-mLWea7o39bcuJundQ6m32tSe3CZoRqVVDAsTbKQKewOhFQwTJzlGjonFNe5d7WeYt_aJ7SJ9O93Zvvc7TbCCGC_82z0_1_JjG1rX67KteRlXYyEQZwxBulB4T7laWqshapf67v96NWmjEdR_FqQfL2gpevdf0b9mnsUfAKpikQ4
CitedBy_id crossref_primary_10_1007_s10994_023_06479_7
crossref_primary_10_1007_s00521_024_10437_2
Cites_doi 10.18653/v1/N16-3020
10.1016/j.media.2015.06.008
10.1145/3387166
10.1109/ICCV.2017.74
10.1609/aaai.v32i1.11833
10.1016/j.inffus.2019.12.012
10.24963/ijcai.2019/184
10.1145/3623377
10.1007/978-3-031-04083-2
10.1109/WCNC49053.2021.9417363
10.5220/0010256208740881
10.1038/nature16961
10.1109/GLOBECOM42002.2020.9322496
10.1007/978-3-030-10928-8_25
10.1109/VTC2020-Fall49728.2020.9348456
10.1007/978-3-319-04717-1
10.1109/INFOCOM48880.2022.9796783
ContentType Journal Article
Copyright 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Copyright_xml – notice: 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
DBID AAYXX
CITATION
ABUWG
AFKRA
AZQEC
BENPR
CCPQU
DWQXO
PHGZM
PHGZT
PIMPY
PKEHL
PQEST
PQQKQ
PQUKI
ADTPV
AFDQA
AOWAS
D8T
D8V
ZZAVC
DOA
DOI 10.3390/app122110947
DatabaseName CrossRef
ProQuest Central (Alumni)
ProQuest Central UK/Ireland
ProQuest Central Essentials - QC
ProQuest Central
ProQuest One
ProQuest Central
ProQuest Central Premium
ProQuest One Academic
Publicly Available Content Database (ProQuest)
ProQuest One Academic Middle East (New)
ProQuest One Academic Eastern Edition (DO NOT USE)
ProQuest One Academic (retired)
ProQuest One Academic UKI Edition
SwePub
SWEPUB Kungliga Tekniska Högskolan full text
SwePub Articles
SWEPUB Freely available online
SWEPUB Kungliga Tekniska Högskolan
SwePub Articles full text
DOAJ Directory of Open Access Journals
DatabaseTitle CrossRef
Publicly Available Content Database
ProQuest Central
ProQuest One Academic Middle East (New)
ProQuest One Academic UKI Edition
ProQuest Central Essentials
ProQuest Central Korea
ProQuest One Academic Eastern Edition
ProQuest Central (Alumni Edition)
ProQuest One Community College
ProQuest Central (New)
ProQuest One Academic
ProQuest One Academic (New)
DatabaseTitleList
CrossRef

Publicly Available Content Database
Database_xml – sequence: 1
  dbid: DOA
  name: DOAJ Directory of Open Access Journals
  url: https://www.doaj.org/
  sourceTypes: Open Website
– sequence: 2
  dbid: PIMPY
  name: Publicly Available Content Database
  url: http://search.proquest.com/publiccontent
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
Sciences (General)
EISSN 2076-3417
ExternalDocumentID oai_doaj_org_article_548b9527aebf492da00b7523c8621c57
oai_DiVA_org_kth_329022
10_3390_app122110947
GroupedDBID .4S
2XV
5VS
7XC
8CJ
8FE
8FG
8FH
AADQD
AAFWJ
AAYXX
ADBBV
ADMLS
AFFHD
AFKRA
AFPKN
AFZYC
ALMA_UNASSIGNED_HOLDINGS
APEBS
ARCSS
BCNDV
BENPR
CCPQU
CITATION
CZ9
D1I
D1J
D1K
GROUPED_DOAJ
IAO
IGS
ITC
K6-
K6V
KC.
KQ8
L6V
LK5
LK8
M7R
MODMG
M~E
OK1
P62
PHGZM
PHGZT
PIMPY
PROAC
TUS
ABUWG
AZQEC
DWQXO
PKEHL
PQEST
PQQKQ
PQUKI
ADTPV
AFDQA
AOWAS
D8T
D8V
IPNFZ
RIG
ZZAVC
ID FETCH-LOGICAL-c405t-488c85a642cf92b56d6b865cf1e3f9dbf3e64782dd4625eb8a87f8fd9713eb6d3
IEDL.DBID PIMPY
ISICitedReferencesCount 3
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000880918200001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 2076-3417
IngestDate Fri Oct 03 12:43:47 EDT 2025
Tue Nov 04 16:36:02 EST 2025
Mon Jun 30 11:23:33 EDT 2025
Sat Nov 29 07:21:45 EST 2025
Tue Nov 18 21:18:38 EST 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 21
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c405t-488c85a642cf92b56d6b865cf1e3f9dbf3e64782dd4625eb8a87f8fd9713eb6d3
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ORCID 0000-0002-0182-8390
0000-0001-7448-3381
0000-0002-6650-2789
OpenAccessLink https://www.proquest.com/publiccontent/docview/2771655104?pq-origsite=%requestingapplication%
PQID 2771655104
PQPubID 2032433
ParticipantIDs doaj_primary_oai_doaj_org_article_548b9527aebf492da00b7523c8621c57
swepub_primary_oai_DiVA_org_kth_329022
proquest_journals_2771655104
crossref_citationtrail_10_3390_app122110947
crossref_primary_10_3390_app122110947
PublicationCentury 2000
PublicationDate 2022-11-01
PublicationDateYYYYMMDD 2022-11-01
PublicationDate_xml – month: 11
  year: 2022
  text: 2022-11-01
  day: 01
PublicationDecade 2020
PublicationPlace Basel
PublicationPlace_xml – name: Basel
PublicationTitle Applied sciences
PublicationYear 2022
Publisher MDPI AG
Publisher_xml – name: MDPI AG
References ref_14
ref_35
ref_12
ref_34
ref_11
ref_33
ref_10
ref_32
ref_31
ref_30
ref_18
ref_17
ref_16
ref_38
ref_15
ref_37
Balcan (ref_13) 2016; Volume 48
ref_25
Silver (ref_2) 2016; 529
ref_24
ref_22
ref_21
ref_20
ref_1
Bennetot (ref_36) 2020; 58
ref_3
Larochelle (ref_19) 2020; Volume 33
Mohseni (ref_23) 2021; 11
ref_29
ref_28
ref_26
ref_8
ref_5
ref_4
Madumal (ref_27) 2020; 34
ref_7
Gaonkar (ref_9) 2015; 24
ref_6
References_xml – ident: ref_7
– ident: ref_28
– ident: ref_24
– ident: ref_34
– ident: ref_5
  doi: 10.18653/v1/N16-3020
– volume: 24
  start-page: 190
  year: 2015
  ident: ref_9
  article-title: Interpreting support vector machine models for multivariate group wise analysis in neuroimaging
  publication-title: Med. Image Anal.
  doi: 10.1016/j.media.2015.06.008
– volume: 11
  start-page: 24
  year: 2021
  ident: ref_23
  article-title: A Multidisciplinary Survey and Framework for Design and Evaluation of Explainable AI Systems
  publication-title: ACM Trans. Interact. Intell. Syst.
  doi: 10.1145/3387166
– ident: ref_37
– ident: ref_14
– ident: ref_1
– ident: ref_11
  doi: 10.1109/ICCV.2017.74
– ident: ref_35
– ident: ref_21
– ident: ref_25
  doi: 10.1609/aaai.v32i1.11833
– volume: 58
  start-page: 82
  year: 2020
  ident: ref_36
  article-title: Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI
  publication-title: Inf. Fusion
  doi: 10.1016/j.inffus.2019.12.012
– ident: ref_26
  doi: 10.24963/ijcai.2019/184
– volume: Volume 33
  start-page: 11298
  year: 2020
  ident: ref_19
  article-title: RD2: Reward Decomposition with Representation Decomposition
  publication-title: Proceedings of the Advances in Neural Information Processing Systems
– ident: ref_18
  doi: 10.1145/3623377
– ident: ref_3
  doi: 10.1007/978-3-031-04083-2
– ident: ref_6
– ident: ref_8
– ident: ref_4
– ident: ref_30
  doi: 10.1109/WCNC49053.2021.9417363
– ident: ref_17
  doi: 10.5220/0010256208740881
– ident: ref_33
– volume: 529
  start-page: 484
  year: 2016
  ident: ref_2
  article-title: Mastering the game of Go with deep neural networks and tree search
  publication-title: Nature
  doi: 10.1038/nature16961
– ident: ref_12
– volume: Volume 48
  start-page: 1899
  year: 2016
  ident: ref_13
  article-title: Graying the black box: Understanding DQNs
  publication-title: Proceedings of the 33rd International Conference on Machine Learning
– volume: 34
  start-page: 2493
  year: 2020
  ident: ref_27
  article-title: Explainable Reinforcement Learning through a Causal Lens
  publication-title: Proc. AAAI Conf. Artif. Intell.
– ident: ref_16
  doi: 10.1109/GLOBECOM42002.2020.9322496
– ident: ref_15
– ident: ref_29
  doi: 10.1007/978-3-030-10928-8_25
– ident: ref_31
  doi: 10.1109/VTC2020-Fall49728.2020.9348456
– ident: ref_38
– ident: ref_10
  doi: 10.1007/978-3-319-04717-1
– ident: ref_22
– ident: ref_20
– ident: ref_32
  doi: 10.1109/INFOCOM48880.2022.9796783
SSID ssj0000913810
Score 2.275952
Snippet Deep Reinforcement Learning (RL) is a black-box method and is hard to understand because the agent employs a neural network (NN). To explain the behavior and...
SourceID doaj
swepub
proquest
crossref
SourceType Open Website
Open Access Repository
Aggregation Database
Enrichment Source
Index Database
StartPage 10947
SubjectTerms Artificial intelligence
Bias
Decomposition
deep reinforcement learning
explainability
explainable reinforcement learning
Methods
Neural networks
Performance evaluation
reward decomposition
reward prioritization
Variables
SummonAdditionalLinks – databaseName: DOAJ Directory of Open Access Journals
  dbid: DOA
  link: http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV1NS8NAEB2keNCD2KpYrbIHFUWCaT53vbXa4qmIqPQW9lOLEqWJ_n5nN2lJD-LFa1iyy5ts5r1k5w3AScioYZjXvNAY5UVGa49Tn3mKYjrXodEJp67ZRDqZ0OmU3TdafdkzYZU9cAXcFTJqweIg5VqYiAWK-75IUT1JpOJ9Gbs6cmQ9DTHl3sGsb62rqpPuIep6-z-4H1i1w2wnlUYOclb9q_yy6Rnq8sx4G7ZqgkgG1cLasKbzDmw2bAM70K43ZEHOa9foix2gwxES02syROjJKFcFscfrePWxryDITcmDdjap0n0RJLWz6ssuPI1Hjzd3Xt0WwZPIrkoPt5ykMUfhIA0LRJyoRNAklsZiy5QwobYFpIFSEYobLSinqaFGMdSjWiQq3INW_pHrfSCcRVzh3WQqMJulVGC25tr4RqsojFTShcsFUJmsPcNt64r3DLWDhTVrwtqF0-Xoz8or45dxQ4v5cox1uHYXMO5ZHffsr7h3obeIWFZvuyILUpR_yAH9qAtnVRRXZrmdPQ_cLG_laxYGDAnMwX8s5hA2AlsW4WoUe9Aq51_6CNbldzkr5sfu0fwBU_Dmaw
  priority: 102
  providerName: Directory of Open Access Journals
Title BEERL: Both Ends Explanations for Reinforcement Learning
URI https://www.proquest.com/docview/2771655104
https://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-329022
https://doaj.org/article/548b9527aebf492da00b7523c8621c57
Volume 12
WOSCitedRecordID wos000880918200001&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: PRVAON
  databaseName: DOAJ Directory of Open Access Journals
  customDbUrl:
  eissn: 2076-3417
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000913810
  issn: 2076-3417
  databaseCode: DOA
  dateStart: 20110101
  isFulltext: true
  titleUrlDefault: https://www.doaj.org/
  providerName: Directory of Open Access Journals
– providerCode: PRVHPJ
  databaseName: ROAD: Directory of Open Access Scholarly Resources
  customDbUrl:
  eissn: 2076-3417
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000913810
  issn: 2076-3417
  databaseCode: M~E
  dateStart: 20110101
  isFulltext: true
  titleUrlDefault: https://road.issn.org
  providerName: ISSN International Centre
– providerCode: PRVPQU
  databaseName: ProQuest Central
  customDbUrl:
  eissn: 2076-3417
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000913810
  issn: 2076-3417
  databaseCode: BENPR
  dateStart: 20110101
  isFulltext: true
  titleUrlDefault: https://www.proquest.com/central
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: Publicly Available Content Database
  customDbUrl:
  eissn: 2076-3417
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000913810
  issn: 2076-3417
  databaseCode: PIMPY
  dateStart: 20110101
  isFulltext: true
  titleUrlDefault: http://search.proquest.com/publiccontent
  providerName: ProQuest
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV1LTxsxEB5B0gMceLWooTTyoa1aVats9mlzQYQGtYdGEWorelr5SREogWzg93fG64RwKKdeV5a92vF4vm_W8w3Au1RwJzCuRalzJsqctZHksYgMx3BuU2cLyX2ziXI04hcXYhzKo-twrXJxJvqDulF7pnvbeAj3zFRTxryXlIjzMdjH2fHtXUQ9pOhfa2iosQ5tEt7iLWiPv30f_17mXEgDk_fj5v57imyf_hL3E-JAgvqrrEQmL-D_FHWuKon66HO2_X_fewe2AgplJ8222YU1O9mDzRVtwj3YDV5fs49BmvrTS-CDIaLfIzZA-7LhxNSM7vDJJqNYMwTA7Nx6LVbt044syLdevoKfZ8Mfp1-j0Hsh0gjh5hH6tea5RHainUhUXphC8SLXjgwojHKppSrVxJgMGZRVXPLScWcEkl6rCpPuQ2syndjXwKTIpMHZdKkwZJZcISSQ1sXOmizNTNGBz4vvXukgTE79MW4qJChkpWrVSh14vxx92why_GPcgEy4HEMy2v7BdHZZBa-skK4pkSeltMplIjEyjlWJ1Fwjz-vrHCc5XBi1Cr5dV4827MCHZlM8WeXL1a8Tv8r1_E-VJgJR0sHz87yBjYSqKnyJ4yG05rN7-xZe6If5VT3rQnswHI3Puz5d0A17-i8sgASx
linkProvider ProQuest
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMw1V1LbxMxEB6VFIlyAFqoCBTwgSIQWrGx92EjIdTQVI3aRhEqqJwWP0sFSko2gPhT_EbGXm9ID3DrgevK8j787cw3Y883AI-Z4E6gX0uYcybJnLWJ5KlIDEd3bpmzheSh2UQ5GvGTEzFegV9tLYw_VtnaxGCozVT7HPkLWiKzR_eeZq_Pvya-a5TfXW1baDSwOLA_f2DIVr8a7uL6blO6Nzh-s5_ErgKJRnIyTxCxmucSebd2gqq8MIXiRa6dfzRhlGPW119SYzKMDazikpeOOyMwnLOqMAznvQKrGYKdd2B1PDwaf1hkdbzKJu-lzQl7xkTq96F71EdZwndwWfJ9oUXARV67rFUa_Nvezf_ty9yCG5FJk50G-uuwYicbcH1JX3ED1qPlqsnTKK_97Dbw_gAZ_EvSR4ySwcTUxJ9DlE1WtCZI4slbG_RkdUidkihBe3oH3l3K-2xCZzKd2LtApMikwdl0qdDtl1whrZHWpc6ajGWm6MLzdmUrHcXVfY-PLxUGWR4H1TIOurC9GH3eiIr8ZVzfg2QxxkuBhwvT2WkVLUuFIacSOS2lVS4T1Mg0VWVOmcZYtadznGSrhU0V7VNd_cFMF540sLtwl92z9zvhLp_nnypGBTK9e_-e5xFc2z8-OqwOh6OD-7BGfZVIKNncgs589s0-gKv6-_ysnj2M_wyBj5cNxt-Lt1T3
linkToPdf http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMw1V1NbxMxEB2VFCE4AC0gAgV8oAiEVt3Y-2EjIdSQRESFKEJQldPiz1KBkpINIP4av46x1xvSA9x64Lqy7N3188wbe_wG4CET3An0awlzziSZszaRPBWJ4ejOLXO2kDwUmygnE350JKYb8Ku9C-PTKlubGAy1mWu_R75HS2T26N7TbM_FtIjpYPTi9GviK0j5k9a2nEYDkQP78weGb_Xz8QDnepfS0fDdy1dJrDCQaCQqywTRq3kukYNrJ6jKC1MoXuTa-dcURjlm_V1MakyGcYJVXPLScWcEhnZWFYZhvxdgEyl5lnVgczp-M_2w2uHxipu8lzbZ9oyJ1J9J96iPuISv5rLmB0O5gLMcd123NPi60bX_-S9dh6uRYZP9ZklswYadbcOVNd3FbdiKFq0mj6Ps9pMbwPtDZPbPSB-xS4YzUxOfnyib3dKaILknb23QmdVhS5VEadrjm_D-XL7nFnRm85m9DUSKTBrsTZcK6UDJFdIdaV3qrMlYZoouPG1nudJRdN3X_vhSYfDlMVGtY6ILu6vWp43YyF_a9T1gVm28RHh4MF8cV9HiVBiKKpHTUlrlMkGNTFNV5pRpjGF7OsdOdloIVdFu1dUf_HThUQPBM6MMTg73wyifl58qRgUywDv_7ucBXEIEVq_Hk4O7cJn6yyPhJucOdJaLb_YeXNTflyf14n5cPgQ-njcWfwPTCl24
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=BEERL%3A+Both+Ends+Explanations+for+Reinforcement+Learning&rft.jtitle=Applied+sciences&rft.au=Terra%2C+Ahmad&rft.au=Inam%2C+Rafia&rft.au=Fersman%2C+Elena&rft.date=2022-11-01&rft.issn=2076-3417&rft.eissn=2076-3417&rft.volume=12&rft.issue=21&rft.spage=10947&rft_id=info:doi/10.3390%2Fapp122110947&rft.externalDBID=n%2Fa&rft.externalDocID=10_3390_app122110947
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2076-3417&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2076-3417&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2076-3417&client=summon