Abstract
Explainable Artificial Intelligence (XAI), i.e., the development of more transparent and interpretable AI models, has gained increased traction over the last few years. This is due to the fact that, in conjunction with their growth into powerful and ubiquitous tools, AI models exhibit one detrimental characteristic: a performance-transparency trade-off. This describes the fact that the more complex a model’s inner workings, the less clear it is how its predictions or decisions were achieved. But, especially considering Machine Learning (ML) methods like Reinforcement Learning (RL) where the system learns autonomously, the necessity to understand the underlying reasoning for their decisions becomes apparent. Since, to the best of our knowledge, there exists no single work offering an overview of Explainable Reinforcement Learning (XRL) methods, this survey attempts to address this gap. We give a short summary of the problem, a definition of important terms, and offer a classification and assessment of current XRL methods. We found that a) the majority of XRL methods function by mimicking and simplifying a complex model instead of designing an inherently simple one, and b) XRL (and XAI) methods often neglect to consider the human side of the equation, not taking into account research from related fields like psychology or philosophy. Thus, an interdisciplinary effort is needed to adapt the generated explanations to a (non-expert) human user in order to effectively progress in the field of XRL and XAI in general.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Please note that, while there is a distinction between Reinforcement Learning and Deep Reinforcement Learning (DRL), for the sake of simplicity, we will refer to both as just Reinforcement Learning going forward.
- 2.
E.g. the AI Explainability 360 (AIX360) as the currently most comprehensive one [4] (see also for a list of other toolkits).
- 3.
- 4.
With the exception of method C in Sect. 3.3, where we present a Linear Model U-Tree method although another paper with a different, but related method was published slightly later. See the last paragraph of that section for our reasoning for this decision.
References
Abdul, A., Vermeulen, J., Wang, D., Lim, B.Y., Kankanhalli, M.: Trends and trajectories for explainable, accountable and intelligible systems. In: Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems - CHI 2018. ACM Press (2018)
Adadi, A., Berrada, M.: Peeking inside the black-box: a survey on explainable artificial intelligence (XAI). IEEE Access 6, 52138–52160 (2018). https://doi.org/10.1109/access.2018.2870052
Andreas, J., Klein, D., Levine, S.: Modular multitask reinforcement learning with policy sketches. In: Proceedings of the 34th International Conference on Machine Learning, ICML 2017, vol. 70, pp. 166–175. JMLR.org (2017)
Arya, V., et al.: One explanation does not fit all: a toolkit and taxonomy of AI explainability techniques (2019). arXiv:1909.03012
Bevana, N., Kirakowskib, J., Maissela, J.: What is usability. In: Proceedings of the 4th International Conference on HCI. Citeseer (1991)
Brockman, G., Cheung, V., Pettersson, L., Schneider, J., Schulman, J., Tang, J., Zaremba, W.: OpenAI gym (2016). arXiv:1606.01540
Carvalho, D.V., Pereira, E.M., Cardoso, J.S.: Machine learning interpretability: a survey on methods and metrics. Electronics 8(8), 832 (2019). https://doi.org/10.3390/electronics8080832
Chakraborty, S., et al.: Interpretability of deep learning models: a survey of results. In: 2017 IEEE SmartWorld, Ubiquitous Intelligence & Computing, Advanced & Trusted Computed, Scalable Computing & Communications, Cloud & Big Data Computing, Internet of People and Smart City Innovation (SmartWorld/SCALCOM/UIC/ATC/CBDCom/IOP/SCI). IEEE (2017)
Coppens, Y., et al.: Distilling deep reinforcement learning policies in soft decision trees. In: Proceedings of the IJCAI 2019 Workshop on Explainable Artificial Intelligence, pp. 1–6 (2019)
Doran, D., Schulz, S., Besold, T.R.: What does explainable AI really mean? A new conceptualization of perspectives (2017). arXiv:1710.00794
Doshi-Velez, F., Kim, B.: Towards a rigorous science of interpretable machine learning (2017). arXiv:1702.08608
Dosilovic, F.K., Brcic, M., Hlupic, N.: Explainable artificial intelligence: a survey. In: 2018 41st International Convention on Information and Communication Technology, Electronics and Microelectronics (MIPRO). IEEE (2018). https://doi.org/10.23919/mipro.2018.840004
Du, M., Liu, N., Hu, X.: Techniques for interpretable machine learning. Commun. ACM 63(1), 68–77 (2019). https://doi.org/10.1145/3359786
European Commission, Parliament: Regulation (EU) 2016/679 of the European parliament and of the council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation). OJ L 119, 1–88 (2016)
Fischer, L., Memmen, J.M., Veith, E.M., Tröschel, M.: Adversarial resilience learning–towards systemic vulnerability analysis for large and complex systems. In: The Ninth International Conference on Smart Grids, Green Communications and IT Energy-aware Technologies (ENERGY 2019), vol. 9, pp. 24–32 (2019)
Freitas, A.A.: Comprehensible classification models. ACM SIGKDD Explor. Newsl. 15(1), 1–10 (2014)
Fukuchi, Y., Osawa, M., Yamakawa, H., Imai, M.: Autonomous self-explanation of behavior for interactive reinforcement learning agents. In: Proceedings of the 5th International Conference on Human Agent Interaction - HAI 2017. ACM Press (2017)
Glass, A., McGuinness, D.L., Wolverton, M.: Toward establishing trust in adaptive agents. In: Proceedings of the 13th International Conference on Intelligent User Interfaces - IUI 2008. ACM Press (2008)
Goodman, B., Flaxman, S.: European union regulations on algorithmic decision-making and a “right to explanation”. AI Mag. 38(3), 50–57 (2017)
Halpern, J.Y.: Causes and explanations: a structural-model approach. Part II: explanations. Br. J. Philos. Sci. 56(4), 889–911 (2005)
Hayes, B., Shah, J.A.: Improving robot controller transparency through autonomous policy explanation. In: Proceedings of the 2017 ACM/IEEE International Conference on Human-Robot Interaction - HRI 2017. ACM Press (2017)
Hein, D., Hentschel, A., Runkler, T., Udluft, S.: Particle swarm optimization for generating interpretable fuzzy reinforcement learning policies. Eng. Appl. Artif. Intell. 65, 87–98 (2017). https://doi.org/10.1016/j.engappai.2017.07.005
Hein, D., Udluft, S., Runkler, T.A.: Interpretable policies for reinforcement learning by genetic programming. Eng. Appl. Artif. Intell. 76, 158–169 (2018)
Herlocker, J.L., Konstan, J.A., Riedl, J.: Explaining collaborative filtering recommendations. In: Proceedings of the 2000 ACM Conference on Computer Supported Cooperative Work - CSCW 2000. ACM Press (2000)
Holzinger, A., Carrington, A., Müller, H.: Measuring the quality of explanations: the system causability scale (SCS). KI - Künstliche Intelligenz 34(2), 193–198 (2020). https://doi.org/10.1007/s13218-020-00636-z
Holzinger, A., Langs, G., Denk, H., Zatloukal, K., Müller, H.: Causability and explainability of artificial intelligence in medicine. WIREs Data Min. Knowl. Disc. 9(4) (2019). https://doi.org/10.1002/widm.1312
Ikonomovska, E., Gama, J., Džeroski, S.: Learning model trees from evolving data streams. Data Min. Knowl. Disc. 23(1), 128–168 (2010)
Israelsen, B.W., Ahmed, N.R.: “Dave...I can assure you...that it’s going to be all right...” a definition, case for, and survey of algorithmic assurances in human-autonomy trust relationships. ACM Comput. Surv. 51(6), 1–37 (2019)
Juozapaitis, Z., Koul, A., Fern, A., Erwig, M., Doshi-Velez, F.: Explainable reinforcement learning via reward decomposition. In: Proceedings of the IJCAI 2019 Workshop on Explainable Artificial Intelligence, pp. 47–53 (2019)
Kaelbling, L.P., Littman, M.L., Moore, A.W.: Reinforcement learning: a survey (1996). arXiv:cs/9605103
Kim, B., Khanna, R., Koyejo, O.O.: Examples are not enough, learn to criticize! criticism for interpretability. In: Lee, D.D., Sugiyama, M., Luxburg, U.V., Guyon, I., Garnett, R. (eds.) Advances in Neural Information Processing Systems 29, pp. 2280–2288. Curran Associates, Inc. (2016). http://papers.nips.cc/paper/6300-examples-are-not-enough-learn-to-criticize-criticism-for-interpretability.pdf
Lee, J.D., See, K.A.: Trust in automation: designing for appropriate reliance. Hum. Fact. J. Hum. Fact. Ergon. Soc. 46(1), 50–80 (2004). https://doi.org/10.1518/hfes.46.1.50_30392
Lee, J.H.: Complementary reinforcement learning towards explainable agents (2019). arXiv:1901.00188
Li, Y.: Deep reinforcement learning (2018). arXiv:1810.06339
Lipton, Z.C.: The mythos of model interpretability (2016). arXiv:1606.03490
Lipton, Z.C.: The mythos of model interpretability. Commun. ACM 61(10), 36–43 (2018)
Liu, G., Schulte, O., Zhu, W., Li, Q.: Toward interpretable deep reinforcement learning with linear model U-Trees. In: Berlingerio, M., Bonchi, F., Gärtner, T., Hurley, N., Ifrim, G. (eds.) ECML PKDD 2018. LNCS (LNAI), vol. 11052, pp. 414–429. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-10928-8_25
Liu, Y., et al.: Detecting cancer metastases on gigapixel pathology images (2017). arXiv:1703.02442
Loh, W.Y.: Classification and regression trees. WIREs Data Min. Knowl. Disc. 1(1), 14–23 (2011)
Madumal, P., Miller, T., Sonenberg, L., Vetere, F.: Explainable reinforcement learning through a causal lens (2019). arXiv:1905.10958
Martens, D., Vanthienen, J., Verbeke, W., Baesens, B.: Performance of classification models from a user perspective. Decis. Support Syst. 51(4), 782–793 (2011)
Miller, T.: Explanation in artificial intelligence: insights from the social sciences. Artif. Intell. 267, 1–38 (2019)
Molar, C.: Interpretable machine learning (2018). https://christophm.github.io/interpretable-ml-book/. Accessed 31 Mar 2020
Montavon, G., Samek, W., Müller, K.R.: Methods for interpreting and understanding deep neural networks. Digit. Signal Proc. 73, 1–15 (2018)
Nguyen, A., Yosinski, J., Clune, J.: Deep neural networks are easily fooled: high confidence predictions for unrecognizable images. In: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2015)
Nguyen, T.T., Hui, P.M., Harper, F.M., Terveen, L., Konstan, J.A.: Exploring the filter bubble. In: Proceedings of the 23rd International Conference on World Wide Web - WWW 2014. ACM Press (2014)
Quinlan, J.R., et al.: Learning with continuous classes. In: 5th Australian Joint Conference on Artificial Intelligence, vol. 92, pp. 343–348. World Scientific (1992)
Ribeiro, M.T., Singh, S., Guestrin, C.: “Why should I trust you?”. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD 2016. ACM Press (2016)
Rudin, C.: Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead. Nat. Mach. Intell. 1(5), 206–215 (2019)
Rusu, A.A., et al.: Policy distillation (2015). arXiv:1511.06295
Schrittwieser, J., et al.: Mastering ATARI, go, chess and shogi by planning with a learned model (2019)
Sequeira, P., Gervasio, M.: Interestingness elements for explainable reinforcement learning: understanding agents’ capabilities and limitations (2019). arXiv:1912.09007
Shu, T., Xiong, C., Socher, R.: Hierarchical and interpretable skill acquisition in multi-task reinforcement learning (2017)
Szegedy, C., et al.: Intriguing properties of neural networks (2013). arXiv:1312.6199
The European Commission: Communication from the Commission to the European Parliament, the European Council, the Council, the European Economic and Social Committee and the Committee of the Regions. The European Commission (2018). https://ec.europa.eu/digital-single-market/en/news/communication-artificial-intelligence-europe. Article. Accessed 27 Mar 2020
The European Commission: Independent High-Level Expert Group on Artificial Intelligence set up by the European Commission. The European Commission (2018). https://ec.europa.eu/digital-single-market/en/news/communication-artificial-intelligence-europe. Article. Accessed 27 Apr 2020
Tomzcak, K., et al.: Let Tesla park your Tesla: driver trust in a semi-automated car. In: 2019 Systems and Information Engineering Design Symposium (SIEDS). IEEE (2019)
Uther, W.T., Veloso, M.M.: Tree based discretization for continuous state space reinforcement learning. In: AAAI/IAAI, pp. 769–774 (1998)
Veith, E., Fischer, L., Tröschel, M., Nieße, A.: Analyzing cyber-physical systems from the perspective of artificial intelligence. In: Proceedings of the 2019 International Conference on Artificial Intelligence, Robotics and Control. ACM (2019)
Veith, E.M.: Universal Smart Grid Agent for Distributed Power Generation Management. Logos Verlag Berlin GmbH, Berlin (2017)
Verma, A., Murali, V., Singh, R., Kohli, P., Chaudhuri, S.: Programmatically interpretable reinforcement learning. PMLR 80, 5045–5054 (2018). arXiv:1804.02477
van der Waa, J., van Diggelen, J., van den Bosch, K., Neerincx, M.: Contrastive explanations for reinforcement learning in terms of expected consequences. In: IJCAI 2018 Workshop on Explainable AI (XAI), vol. 37 (2018). arXiv:1807.08706
Wymann, B., Espié, E., Guionneau, C., Dimitrakakis, C., Coulom, R., Sumner, A.: TORCS, the open racing car simulator, vol. 4, no. 6, p. 2 (2000). Software http://torcs.sourceforge.net
Zahavy, T., Zrihem, N.B., Mannor, S.: Graying the black box: understanding DQNs (2016). arXiv:1602.02658
Zhou, J., Chen, F. (eds.): Human and Machine Learning. HIS. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-90403-0
Zhou, J., Chen, F.: Towards trustworthy human-AI teaming under uncertainty. In: IJCAI 2019 Workshop on Explainable AI (XAI) (2019)
Zhou, J., Hu, H., Li, Z., Yu, K., Chen, F.: Physiological indicators for user trust in machine learning with influence enhanced fact-checking. In: Holzinger, A., Kieseberg, P., Tjoa, A.M., Weippl, E. (eds.) CD-MAKE 2019. LNCS, vol. 11713, pp. 94–113. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-29726-8_7
Acknowledgements
This work was supported by the German Research Foundation under the grant GZ: JI 140/7-1. We thank our colleagues Stephan Balduin, Johannes Gerster, Lasse Hammer, Daniel Lange and Nils Wenninghoff for their helpful comments and contributions.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 IFIP International Federation for Information Processing
About this paper
Cite this paper
Puiutta, E., Veith, E.M.S.P. (2020). Explainable Reinforcement Learning: A Survey. In: Holzinger, A., Kieseberg, P., Tjoa, A., Weippl, E. (eds) Machine Learning and Knowledge Extraction. CD-MAKE 2020. Lecture Notes in Computer Science(), vol 12279. Springer, Cham. https://doi.org/10.1007/978-3-030-57321-8_5
Download citation
DOI: https://doi.org/10.1007/978-3-030-57321-8_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-57320-1
Online ISBN: 978-3-030-57321-8
eBook Packages: Computer ScienceComputer Science (R0)