Model tree methods for explaining deep reinforcement learning agents in real-time robotic applications

Deep reinforcement learning has shown useful in the field of robotics but the black-box nature of deep neural networks impedes the applicability of deep reinforcement learning agents for real-world tasks. This is addressed in the field of explainable artificial intelligence, by developing explanatio...

Full description

Saved in:
Bibliographic Details
Published in:Neurocomputing (Amsterdam) Vol. 515; pp. 133 - 144
Main Authors: Gjærum, Vilde B., Strümke, Inga, Løver, Jakob, Miller, Timothy, Lekkas, Anastasios M.
Format: Journal Article
Language:English
Published: Elsevier B.V 01.01.2023
Subjects:
ISSN:0925-2312, 1872-8286
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract Deep reinforcement learning has shown useful in the field of robotics but the black-box nature of deep neural networks impedes the applicability of deep reinforcement learning agents for real-world tasks. This is addressed in the field of explainable artificial intelligence, by developing explanation methods that aim to explain such agents to humans. Model trees as surrogate models have proven useful for producing explanations for black-box models used in real-world robotic applications, in particular, due to their capability of providing explanations in real time. In this paper, we provide an overview and analysis of available methods for building model trees for explaining deep reinforcement learning agents solving robotics tasks. We find that multiple outputs are important for the model to be able to grasp the dependencies of coupled output features, i.e.actions. Additionally, our results indicate that introducing domain knowledge via a hierarchy among the input features during the building process results in higher accuracies and a faster building process.
AbstractList Deep reinforcement learning has shown useful in the field of robotics but the black-box nature of deep neural networks impedes the applicability of deep reinforcement learning agents for real-world tasks. This is addressed in the field of explainable artificial intelligence, by developing explanation methods that aim to explain such agents to humans. Model trees as surrogate models have proven useful for producing explanations for black-box models used in real-world robotic applications, in particular, due to their capability of providing explanations in real time. In this paper, we provide an overview and analysis of available methods for building model trees for explaining deep reinforcement learning agents solving robotics tasks. We find that multiple outputs are important for the model to be able to grasp the dependencies of coupled output features, i.e.actions. Additionally, our results indicate that introducing domain knowledge via a hierarchy among the input features during the building process results in higher accuracies and a faster building process.
Author Lekkas, Anastasios M.
Strümke, Inga
Miller, Timothy
Gjærum, Vilde B.
Løver, Jakob
Author_xml – sequence: 1
  givenname: Vilde B.
  surname: Gjærum
  fullname: Gjærum, Vilde B.
  email: vilde.gjarum@ntnu.no
  organization: Department of Engineering Cybernetics, Norwegian University of Science and Technology, 7034 Trondheim, Norway
– sequence: 2
  givenname: Inga
  surname: Strümke
  fullname: Strümke, Inga
  email: inga.strumke@ntnu.no
  organization: Department of Computer Science, Norwegian University of Science and Technology, 7034 Trondheim, Norway
– sequence: 3
  givenname: Jakob
  surname: Løver
  fullname: Løver, Jakob
  email: loverjakob@gmail.com
  organization: Department of Engineering Cybernetics, Norwegian University of Science and Technology, 7052 Trondheim, Norway
– sequence: 4
  givenname: Timothy
  surname: Miller
  fullname: Miller, Timothy
  email: tmiller@unimelb.edu.au
  organization: School of Computing and Information Systems, University of Melbourne, 3010 Melbourne, Australia
– sequence: 5
  givenname: Anastasios M.
  surname: Lekkas
  fullname: Lekkas, Anastasios M.
  email: tmiller@unimelb.edu.au
  organization: Department of Engineering Cybernetics, Norwegian University of Science and Technology, 7034 Trondheim, Norway
BookMark eNqFkE1LAzEQhoNUsK3-Aw_5A7vmaz_qQZCiVah40XNIk9maspssSRT996atJw96GuadeV5m3hmaOO8AoUtKSkpofbUrHbxrP5SMMJalklBxgqa0bVjRsraeoClZsKpgnLIzNItxRwhtKFtMUffkDfQ4BQA8QHrzJuLOBwyfY6-ss26LDcCIA1iXdQ0DuIR7UOEwU9vcRmxdXlB9kewAOPiNT1ZjNY691SpZ7-I5Ou1UH-Hip87R6_3dy_KhWD-vHpe360LziqVCkZrRWtRtYwTnZFFx0lS82ihiOHDGRdVoUgsulOCUNNrwLGvK2rbtasY0nyNx9NXBxxigk2OwgwpfkhK5z0ru5DEruc9qr-asMnb9C9M2HS5PQdn-P_jmCEN-7MNCkFFbcBqMDaCTNN7-bfANOCWKTg
CitedBy_id crossref_primary_10_1109_ACCESS_2023_3335912
crossref_primary_10_1109_LRA_2024_3389813
crossref_primary_10_1016_j_egyr_2024_05_031
crossref_primary_10_1016_j_neucom_2025_130642
crossref_primary_10_3390_electronics13071169
crossref_primary_10_1007_s00521_024_10437_2
crossref_primary_10_1177_17298806241254336
crossref_primary_10_1016_j_oceaneng_2024_117067
crossref_primary_10_1016_j_neucom_2023_126790
crossref_primary_10_1016_j_neucom_2024_127591
crossref_primary_10_1016_j_neucom_2025_131061
crossref_primary_10_1016_j_apor_2024_103977
Cites_doi 10.1109/ACCESS.2018.2870052
10.24963/ijcai.2020/662
10.1609/aaai.v33i01.33011624
10.1016/j.ifacol.2021.10.086
10.1016/j.orl.2021.01.002
10.1609/aaai.v34i04.5711
10.1145/3236386.3241340
10.1609/aaai.v32i1.11491
10.1609/aaai.v35i5.16509
10.1145/2939672.2939778
10.1109/MIS.2019.2957223
10.1609/aaai.v34i04.5717
10.1016/j.inffus.2019.12.012
10.18637/jss.v033.i01
ContentType Journal Article
Copyright 2022 The Author(s)
Copyright_xml – notice: 2022 The Author(s)
DBID 6I.
AAFTH
AAYXX
CITATION
DOI 10.1016/j.neucom.2022.10.014
DatabaseName ScienceDirect Open Access Titles
Elsevier:ScienceDirect:Open Access
CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Computer Science
EISSN 1872-8286
EndPage 144
ExternalDocumentID 10_1016_j_neucom_2022_10_014
S0925231222012735
GroupedDBID ---
--K
--M
.DC
.~1
0R~
123
1B1
1~.
1~5
4.4
457
4G.
53G
5VS
6I.
7-5
71M
8P~
9JM
9JN
AABNK
AACTN
AADPK
AAEDT
AAEDW
AAFTH
AAIAV
AAIKJ
AAKOC
AALRI
AAOAW
AAQFI
AAXLA
AAXUO
AAYFN
ABBOA
ABCQJ
ABFNM
ABJNI
ABMAC
ABYKQ
ACDAQ
ACGFS
ACRLP
ACZNC
ADBBV
ADEZE
AEBSH
AEKER
AENEX
AFKWA
AFTJW
AFXIZ
AGHFR
AGUBO
AGWIK
AGYEJ
AHHHB
AHZHX
AIALX
AIEXJ
AIKHN
AITUG
AJOXV
ALMA_UNASSIGNED_HOLDINGS
AMFUW
AMRAJ
AOUOD
AXJTR
BKOJK
BLXMC
CS3
DU5
EBS
EFJIC
EFLBG
EO8
EO9
EP2
EP3
F5P
FDB
FIRID
FNPLU
FYGXN
G-Q
GBLVA
GBOLZ
IHE
J1W
KOM
LG9
M41
MO0
MOBAO
N9A
O-L
O9-
OAUVE
OZT
P-8
P-9
P2P
PC.
Q38
ROL
RPZ
SDF
SDG
SDP
SES
SPC
SPCBC
SSN
SSV
SSZ
T5K
ZMT
~G-
29N
9DU
AAQXK
AATTM
AAXKI
AAYWO
AAYXX
ABWVN
ABXDB
ACLOT
ACNNM
ACRPL
ACVFH
ADCNI
ADJOM
ADMUD
ADNMO
AEIPS
AEUPX
AFJKZ
AFPUW
AGQPQ
AIGII
AIIUN
AKBMS
AKRWK
AKYEP
ANKPU
APXCP
ASPBG
AVWKF
AZFZN
CITATION
EFKBS
EJD
FEDTE
FGOYB
HLZ
HVGLF
HZ~
R2-
SBC
SEW
WUQ
XPP
~HD
ID FETCH-LOGICAL-c352t-a062164687d433095307535ba0d3e323457c06434a43107cd33e3c12888f622c3
ISICitedReferencesCount 14
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000877611700007&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0925-2312
IngestDate Tue Nov 18 20:38:48 EST 2025
Sat Nov 29 07:19:57 EST 2025
Fri Feb 23 02:42:02 EST 2024
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Keywords Explainable artificial intelligence
Model trees
Robotics
Reinforcement learning
Language English
License This is an open access article under the CC BY license.
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c352t-a062164687d433095307535ba0d3e323457c06434a43107cd33e3c12888f622c3
OpenAccessLink https://dx.doi.org/10.1016/j.neucom.2022.10.014
PageCount 12
ParticipantIDs crossref_primary_10_1016_j_neucom_2022_10_014
crossref_citationtrail_10_1016_j_neucom_2022_10_014
elsevier_sciencedirect_doi_10_1016_j_neucom_2022_10_014
PublicationCentury 2000
PublicationDate 2023-01-01
2023-01-00
PublicationDateYYYYMMDD 2023-01-01
PublicationDate_xml – month: 01
  year: 2023
  text: 2023-01-01
  day: 01
PublicationDecade 2020
PublicationTitle Neurocomputing (Amsterdam)
PublicationYear 2023
Publisher Elsevier B.V
Publisher_xml – name: Elsevier B.V
References Adadi, Berrada (b0005) 2018; 6
A.B. Arrieta, N. Díaz-Rodríguez, J. Del Ser, A. Bennetot, S. Tabik, A. Barbado, S. Garcia, Salvador ang Gil-Lopez, D. Molina, R. Benjamins, R. Chatila, and F. Herrera. Explainable artificial intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI. Information Fusion, 58:82–115, 2020.
J. Schulman, F. Wolski, P. Dhariwal, A. Radford, and O. Klimov. Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347, 2017.
Bertsimas, Dunn, Wang (b0030) 2021; 49
Friedman, Hastie, Tibshirani (b0045) 2010; 33
Remman, Lekkas (b0100) 2021
Gjærum, Strümke, Alsos, Lekkas (b0055) 2021; 8
Bertsimas, Dunn (b0025) 2019; LLC
S.M. Lundberg and S.-I. Lee. A unified approach to interpreting model predictions. In Proceedings of the 31st International Conference on Neural Information Processing Systems (NeurIPS), page 4768–4777, 2017.
Singh, Hartikainen, Levine (b0140) 2019
S. Verwer and Y. Zhang. Learning optimal classification trees using a binary linear program formulation. Proceedings of the AAAI Conference on Artificial Intelligence, 33(01), 1625–1632, Jul. 2019.
Haarnoja, Ha, Zhou, Tan, Tucker, Levine (b0070) 2019
Breiman, Friedman, Olshen, Stone (b0035) 1984
I. Covert, S. Lundberg, and S.-I. Lee. Understanding global feature contributions with additive importance measures. 4th Conference on Neural Information Processing Systems (NeurIPS), 2020.
Gjærum, Rørvik, Lekkas (b0050) 2021
G. Aglin, S. Nijssen, and P. Schaus. Learning optimal decision trees using caching branch-and-bound search. Proceedings of the AAAI Conference on Artificial Intelligence, 34(04), 3146–3153, 2020.
A. Silva, T. Killian, I.D.J. Rodriguez, S.-H. Son, and M. Gombolay. Optimization methods for interpretable differentiable decision trees in reinforcement learning. Proceedings of the 23rd International Conference on Artificial Intelligence and Statistics (AISTATS), 2020.
T. Lillicrap, J.J. Hunt, A. Pritzel, N. Heess, T. Erez, Y. Tassa, D. Silver, and D. Wierstra. Continuous control with deep reinforcement learning. 4th International Conference on Learning Representations(ICLR), 2016.
Rørvik (b0115) 2020
J.A. Glomsrud, A. Ødegårdstuen, A.L.S. Clair, and O. Smogeli. Trustworthy versus explainable AI in autonomous vessels. International Seminar on Safety and Security of Autonomous Vessels (ISSAV), 2019.
M.T. Ribeiro, S. Singh, and C. Guestrin. why should i trust you?: Explaining the predictions of any classifier. Proceedings of the 22nd International Conference on Knowledge Discovery and Data Mining, 2016.
M. Sundararajan, A. Taly, and Q. Yan. Axiomatic attribution for deep networks. In Proceedings of the 34th International Conference on Machine Learning, volume 70, page 3319–3328, 2017.
Lipton (b0080) 2018; 16
J. Løver, V.B. Gjærum, and A.M. Lekkas. Explainable AI methods on a deep reinforcement learning agent for automatic docking. IFAC-PapersOnLine, 54(16):146–152, 2021. 13th IFAC Conference on Control Applications in Marine Systems, Robotics, and Vehicles (CAMS).
J. Schulman, P. Moritz, S. Levine, M.I. Jordan, and P. Abbeel. High-dimensional continuous control using generalized advantage estimation. International Conference on learning representation (ICLR), 2016.
J. van der Waa, M. Robeer, J. van Diggelen, M. Brinkhuis, and M. Neerincx. Contrastive explanations with local foil trees. arXiv preprint arXiv:1806.07470, 2018.
A. Schidler and S. Szeider. SAT-based decision tree learning for large data sets. Proceedings of the 35th AAAI Conference on Artificial Intelligence, (5):3904–3912, 2021.
M.T. Ribeiro, S. Singh, and C. Guestrin. Anchors: High-precision model-agnostic explanations. Proceedings of the 32nd AAAI Conference on Artificial Intelligence, 2018.
F. Avellaneda. Efficient inference of optimal decision trees. Proceedings of the AAAI Conference on Artificial Intelligence, 34(04), 3195–3202, 2020.
Løver (b0090) 2021
H. Verhaeghe, S. Nijssen, G. Pesant, C.-G. Quimper, and P. Schaus. Learning optimal decision trees using constraint programming. In Proceedings of the 29th International Joint Conference on Artificial Intelligence (IJCAI), pages 4765–4769, 2020.
Guidotti, Monreale, Giannotti, Pedreschi, Ruggieri, Turini (b0065) 2019; 34
Remman (10.1016/j.neucom.2022.10.014_b0100) 2021
10.1016/j.neucom.2022.10.014_b0015
10.1016/j.neucom.2022.10.014_b0155
10.1016/j.neucom.2022.10.014_b0110
Gjærum (10.1016/j.neucom.2022.10.014_b0050) 2021
10.1016/j.neucom.2022.10.014_b0135
Adadi (10.1016/j.neucom.2022.10.014_b0005) 2018; 6
Friedman (10.1016/j.neucom.2022.10.014_b0045) 2010; 33
Løver (10.1016/j.neucom.2022.10.014_b0090) 2021
10.1016/j.neucom.2022.10.014_b0095
10.1016/j.neucom.2022.10.014_b0150
10.1016/j.neucom.2022.10.014_b0010
10.1016/j.neucom.2022.10.014_b0075
10.1016/j.neucom.2022.10.014_b0130
Bertsimas (10.1016/j.neucom.2022.10.014_b0030) 2021; 49
Breiman (10.1016/j.neucom.2022.10.014_b0035) 1984
10.1016/j.neucom.2022.10.014_b0125
Guidotti (10.1016/j.neucom.2022.10.014_b0065) 2019; 34
10.1016/j.neucom.2022.10.014_b0105
Singh (10.1016/j.neucom.2022.10.014_b0140) 2019
10.1016/j.neucom.2022.10.014_b0145
Lipton (10.1016/j.neucom.2022.10.014_b0080) 2018; 16
10.1016/j.neucom.2022.10.014_b0085
Rørvik (10.1016/j.neucom.2022.10.014_b0115) 2020
10.1016/j.neucom.2022.10.014_b0040
10.1016/j.neucom.2022.10.014_b0120
10.1016/j.neucom.2022.10.014_b0020
Haarnoja (10.1016/j.neucom.2022.10.014_b0070) 2019
Gjærum (10.1016/j.neucom.2022.10.014_b0055) 2021; 8
10.1016/j.neucom.2022.10.014_b0160
10.1016/j.neucom.2022.10.014_b0060
Bertsimas (10.1016/j.neucom.2022.10.014_b0025) 2019; LLC
References_xml – volume: 6
  start-page: 52138
  year: 2018
  end-page: 52160
  ident: b0005
  article-title: Peeking inside the black-box: A survey on explainable artificial intelligence (XAI)
  publication-title: IEEE Access
– reference: M.T. Ribeiro, S. Singh, and C. Guestrin. Anchors: High-precision model-agnostic explanations. Proceedings of the 32nd AAAI Conference on Artificial Intelligence, 2018.
– reference: M.T. Ribeiro, S. Singh, and C. Guestrin. why should i trust you?: Explaining the predictions of any classifier. Proceedings of the 22nd International Conference on Knowledge Discovery and Data Mining, 2016.
– year: 2019
  ident: b0140
  article-title: End-to-end robotic reinforcement learning without reward engineering
  publication-title: Robot.: Sci. Syst. (RSS)
– reference: S.M. Lundberg and S.-I. Lee. A unified approach to interpreting model predictions. In Proceedings of the 31st International Conference on Neural Information Processing Systems (NeurIPS), page 4768–4777, 2017.
– reference: F. Avellaneda. Efficient inference of optimal decision trees. Proceedings of the AAAI Conference on Artificial Intelligence, 34(04), 3195–3202, 2020.
– year: 2019
  ident: b0070
  article-title: Learning to walk via deep reinforcement learning
  publication-title: Robotics: Sci. Syst. (RSS)
– volume: 49
  start-page: 201
  year: 2021
  end-page: 206
  ident: b0030
  article-title: Near-optimal nonlinear regression trees
  publication-title: Oper. Res. Lett.
– reference: I. Covert, S. Lundberg, and S.-I. Lee. Understanding global feature contributions with additive importance measures. 4th Conference on Neural Information Processing Systems (NeurIPS), 2020.
– reference: A. Silva, T. Killian, I.D.J. Rodriguez, S.-H. Son, and M. Gombolay. Optimization methods for interpretable differentiable decision trees in reinforcement learning. Proceedings of the 23rd International Conference on Artificial Intelligence and Statistics (AISTATS), 2020.
– reference: H. Verhaeghe, S. Nijssen, G. Pesant, C.-G. Quimper, and P. Schaus. Learning optimal decision trees using constraint programming. In Proceedings of the 29th International Joint Conference on Artificial Intelligence (IJCAI), pages 4765–4769, 2020.
– year: 2021
  ident: b0100
  article-title: Robotic lever manipulation using hindsight experience replay and shapley additive explanations
  publication-title: European Control Conference (ECC)
– reference: J. Schulman, F. Wolski, P. Dhariwal, A. Radford, and O. Klimov. Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347, 2017.
– reference: T. Lillicrap, J.J. Hunt, A. Pritzel, N. Heess, T. Erez, Y. Tassa, D. Silver, and D. Wierstra. Continuous control with deep reinforcement learning. 4th International Conference on Learning Representations(ICLR), 2016.
– year: 2020
  ident: b0115
  article-title: Automatic docking of an autonomous surface vessel: Developed using deep reinforcement learning and analysed with Explainable AI
– year: 1984
  ident: b0035
  article-title: Classification and regression trees
– volume: 16
  start-page: 31
  year: 2018
  end-page: 57
  ident: b0080
  article-title: The mythos of model interpretability: In machine learning, the concept of interpretability is both important and slippery
  publication-title: ACM Queue
– volume: LLC
  year: 2019
  ident: b0025
  article-title: Machine learning under a modern optimization lens
  publication-title: Dynamic Ideas
– reference: G. Aglin, S. Nijssen, and P. Schaus. Learning optimal decision trees using caching branch-and-bound search. Proceedings of the AAAI Conference on Artificial Intelligence, 34(04), 3146–3153, 2020.
– reference: M. Sundararajan, A. Taly, and Q. Yan. Axiomatic attribution for deep networks. In Proceedings of the 34th International Conference on Machine Learning, volume 70, page 3319–3328, 2017.
– volume: 33
  start-page: 1
  year: 2010
  end-page: 22
  ident: b0045
  article-title: Regularization paths for generalized linear models via coordinate descent
  publication-title: J. Stat. Softw.
– year: 2021
  ident: b0050
  article-title: Approximating a deep reinforcement learning docking agent using linear model trees
  publication-title: European Control Conference(ECC)
– volume: 8
  year: 2021
  ident: b0055
  article-title: Explaining a deep reinforcement learning docking agent using linear model trees and user adapted visualizations
  publication-title: J. Mar. Sci. Eng. (JMSE)
– volume: 34
  start-page: 14
  year: 2019
  end-page: 23
  ident: b0065
  article-title: Factual and counterfactual explanations for black box decision making
  publication-title: IEEE Intell. Syst.
– year: 2021
  ident: b0090
  article-title: Explaining a deep reinforcement learning agent using regression trees
– reference: J.A. Glomsrud, A. Ødegårdstuen, A.L.S. Clair, and O. Smogeli. Trustworthy versus explainable AI in autonomous vessels. International Seminar on Safety and Security of Autonomous Vessels (ISSAV), 2019.
– reference: S. Verwer and Y. Zhang. Learning optimal classification trees using a binary linear program formulation. Proceedings of the AAAI Conference on Artificial Intelligence, 33(01), 1625–1632, Jul. 2019.
– reference: A.B. Arrieta, N. Díaz-Rodríguez, J. Del Ser, A. Bennetot, S. Tabik, A. Barbado, S. Garcia, Salvador ang Gil-Lopez, D. Molina, R. Benjamins, R. Chatila, and F. Herrera. Explainable artificial intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI. Information Fusion, 58:82–115, 2020.
– reference: A. Schidler and S. Szeider. SAT-based decision tree learning for large data sets. Proceedings of the 35th AAAI Conference on Artificial Intelligence, (5):3904–3912, 2021.
– reference: J. Schulman, P. Moritz, S. Levine, M.I. Jordan, and P. Abbeel. High-dimensional continuous control using generalized advantage estimation. International Conference on learning representation (ICLR), 2016.
– reference: J. Løver, V.B. Gjærum, and A.M. Lekkas. Explainable AI methods on a deep reinforcement learning agent for automatic docking. IFAC-PapersOnLine, 54(16):146–152, 2021. 13th IFAC Conference on Control Applications in Marine Systems, Robotics, and Vehicles (CAMS).
– reference: J. van der Waa, M. Robeer, J. van Diggelen, M. Brinkhuis, and M. Neerincx. Contrastive explanations with local foil trees. arXiv preprint arXiv:1806.07470, 2018.
– volume: 6
  start-page: 52138
  year: 2018
  ident: 10.1016/j.neucom.2022.10.014_b0005
  article-title: Peeking inside the black-box: A survey on explainable artificial intelligence (XAI)
  publication-title: IEEE Access
  doi: 10.1109/ACCESS.2018.2870052
– year: 2021
  ident: 10.1016/j.neucom.2022.10.014_b0090
– ident: 10.1016/j.neucom.2022.10.014_b0155
  doi: 10.24963/ijcai.2020/662
– ident: 10.1016/j.neucom.2022.10.014_b0160
  doi: 10.1609/aaai.v33i01.33011624
– ident: 10.1016/j.neucom.2022.10.014_b0040
– ident: 10.1016/j.neucom.2022.10.014_b0095
  doi: 10.1016/j.ifacol.2021.10.086
– ident: 10.1016/j.neucom.2022.10.014_b0135
– volume: 49
  start-page: 201
  year: 2021
  ident: 10.1016/j.neucom.2022.10.014_b0030
  article-title: Near-optimal nonlinear regression trees
  publication-title: Oper. Res. Lett.
  doi: 10.1016/j.orl.2021.01.002
– volume: LLC
  year: 2019
  ident: 10.1016/j.neucom.2022.10.014_b0025
  article-title: Machine learning under a modern optimization lens
  publication-title: Dynamic Ideas
– year: 2020
  ident: 10.1016/j.neucom.2022.10.014_b0115
– ident: 10.1016/j.neucom.2022.10.014_b0010
  doi: 10.1609/aaai.v34i04.5711
– volume: 16
  start-page: 31
  issue: 3
  year: 2018
  ident: 10.1016/j.neucom.2022.10.014_b0080
  article-title: The mythos of model interpretability: In machine learning, the concept of interpretability is both important and slippery
  publication-title: ACM Queue
  doi: 10.1145/3236386.3241340
– year: 1984
  ident: 10.1016/j.neucom.2022.10.014_b0035
– ident: 10.1016/j.neucom.2022.10.014_b0125
– ident: 10.1016/j.neucom.2022.10.014_b0130
– ident: 10.1016/j.neucom.2022.10.014_b0075
– volume: 8
  issue: 9
  year: 2021
  ident: 10.1016/j.neucom.2022.10.014_b0055
  article-title: Explaining a deep reinforcement learning docking agent using linear model trees and user adapted visualizations
  publication-title: J. Mar. Sci. Eng. (JMSE)
– ident: 10.1016/j.neucom.2022.10.014_b0110
  doi: 10.1609/aaai.v32i1.11491
– ident: 10.1016/j.neucom.2022.10.014_b0120
  doi: 10.1609/aaai.v35i5.16509
– year: 2021
  ident: 10.1016/j.neucom.2022.10.014_b0050
  article-title: Approximating a deep reinforcement learning docking agent using linear model trees
  publication-title: European Control Conference(ECC)
– ident: 10.1016/j.neucom.2022.10.014_b0085
– ident: 10.1016/j.neucom.2022.10.014_b0105
  doi: 10.1145/2939672.2939778
– year: 2019
  ident: 10.1016/j.neucom.2022.10.014_b0140
  article-title: End-to-end robotic reinforcement learning without reward engineering
  publication-title: Robot.: Sci. Syst. (RSS)
– ident: 10.1016/j.neucom.2022.10.014_b0060
– volume: 34
  start-page: 14
  issue: 6
  year: 2019
  ident: 10.1016/j.neucom.2022.10.014_b0065
  article-title: Factual and counterfactual explanations for black box decision making
  publication-title: IEEE Intell. Syst.
  doi: 10.1109/MIS.2019.2957223
– year: 2019
  ident: 10.1016/j.neucom.2022.10.014_b0070
  article-title: Learning to walk via deep reinforcement learning
  publication-title: Robotics: Sci. Syst. (RSS)
– ident: 10.1016/j.neucom.2022.10.014_b0020
  doi: 10.1609/aaai.v34i04.5717
– year: 2021
  ident: 10.1016/j.neucom.2022.10.014_b0100
  article-title: Robotic lever manipulation using hindsight experience replay and shapley additive explanations
  publication-title: European Control Conference (ECC)
– ident: 10.1016/j.neucom.2022.10.014_b0150
– ident: 10.1016/j.neucom.2022.10.014_b0015
  doi: 10.1016/j.inffus.2019.12.012
– volume: 33
  start-page: 1
  issue: 1
  year: 2010
  ident: 10.1016/j.neucom.2022.10.014_b0045
  article-title: Regularization paths for generalized linear models via coordinate descent
  publication-title: J. Stat. Softw.
  doi: 10.18637/jss.v033.i01
– ident: 10.1016/j.neucom.2022.10.014_b0145
SSID ssj0017129
Score 2.4450185
Snippet Deep reinforcement learning has shown useful in the field of robotics but the black-box nature of deep neural networks impedes the applicability of deep...
SourceID crossref
elsevier
SourceType Enrichment Source
Index Database
Publisher
StartPage 133
SubjectTerms Explainable artificial intelligence
Model trees
Reinforcement learning
Robotics
Title Model tree methods for explaining deep reinforcement learning agents in real-time robotic applications
URI https://dx.doi.org/10.1016/j.neucom.2022.10.014
Volume 515
WOSCitedRecordID wos000877611700007&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: 1872-8286
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0017129
  issn: 0925-2312
  databaseCode: AIEXJ
  dateStart: 19950101
  isFulltext: true
  titleUrlDefault: https://www.sciencedirect.com
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV3fb9MwELbKxgMv_EaMAfIDb5Grxo7j5LGbBmxCExID9S1KHLdql6VV1k39t_gPd2c7WVgRvyReosptWsv39fzlfPcdIe94nCTTJDIsykeCRUVpWGFkwaYyRG0ZgypWttmEOj1NJpP082Dwva2Fua5UXSebTbr6r6aGMTA2ls7-hbm7L4UBeA1GhyuYHa5_ZHjsbmYTyI1vD33pZL03q8p1gwhKY1ZBY6xmqrbhwbZ5xCzIZ7bkzVa55BXD1vNBsyyWVte1d9jdJ7VW4EPb9hA-8DC-QP2FEsHWBRo-LOyhfNy4zsrf5lVpgoNhF-JZ20P7g8MLly50XM-6HeOTfSfxyaYn-fmy6IDS1TJ61PXDGFzcCWNs19e4ICWXDBio89fGuehEcVv83vfhMpQ9Lxw6bQ2_oYdOYHJrr3Bhi8WwNleYOAST4kNM9HNFrXdUuL_gVHAmnONhvZD3yC5XMgVHujs-PpqcdEdXKuRO4NFPva3XtEmF27_1cz7U4zhnj8lD_3BCxw5UT8jA1E_Jo7bxB_X7wDMytRijiDHqMUYBS_QWYxQxRn_AGG0xRh3G6LymHcaoxxjtY-w5-fr-6OzwI_MNO5gGHr9m-SjmqFeXqDISApUMgZAKWeSjUhjBRSSVRgoc5UBbR0qXAoY1MCTwFzHnWrwgO_WyNi8JjVMT5aHWsIooySbSMDWhiWXBYTSdhntEtMuWaa9mj01VqqxNW1xkbrEzXGwchcXeI6y7a-XUXH7zedVaJPOM1DHNDED0yztf_fOd--TB7f_jNdlZN1fmDbmvr9fzy-atR9sN3zawPQ
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=Model+tree+methods+for+explaining+deep+reinforcement+learning+agents+in+real-time+robotic+applications&rft.jtitle=Neurocomputing+%28Amsterdam%29&rft.au=Gj%C3%A6rum%2C+Vilde+B.&rft.au=Str%C3%BCmke%2C+Inga&rft.au=L%C3%B8ver%2C+Jakob&rft.au=Miller%2C+Timothy&rft.date=2023-01-01&rft.pub=Elsevier+B.V&rft.issn=0925-2312&rft.eissn=1872-8286&rft.volume=515&rft.spage=133&rft.epage=144&rft_id=info:doi/10.1016%2Fj.neucom.2022.10.014&rft.externalDocID=S0925231222012735
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0925-2312&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0925-2312&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0925-2312&client=summon