Neurosymbolic Reinforcement Learning with Formally Verified Exploration
作者： Isil Dillig , Swarat Chaudhuri , Abhinav Verma , Greg Anderson
DOI:
关键词: Reinforcement learning 、 Space (commercial competition) 、 Artificial neural network 、 Class (computer programming) 、 LOOP (programming language) 、 Computer science 、 Key (cryptography) 、 Action (philosophy) 、 State (computer science) 、 Artificial intelligence
摘要: We present REVEL, a partially neural reinforcement learning (RL) framework for provably safe exploration in continuous state and action spaces. A key challenge for provably safe …
arxiv.org参考文章(28)
Fernando Fernández, Javier García, A comprehensive survey on safe reinforcement learning Journal of Machine Learning Research. ,vol. 16, pp. 1437- 1480 ,(2015)
Jeremy H. Gillula, Claire J. Tomlin, Guaranteed Safe Online Learning via Reachability: tracking a ground target using a quadrotor international conference on robotics and automation. pp. 2723- 2730 ,(2012) , 10.1109/ICRA.2012.6225136
Patrick Cousot, Radhia Cousot, Abstract interpretation Proceedings of the 4th ACM SIGACT-SIGPLAN symposium on Principles of programming languages - POPL '77. pp. 238- 252 ,(1977) , 10.1145/512950.512973
Anayo K. Akametalu, Jaime F. Fisac, Jeremy H. Gillula, Shahab Kaynama, Melanie N. Zeilinger, Claire J. Tomlin, Reachability-based safe learning with Gaussian processes conference on decision and control. pp. 1424- 1431 ,(2014) , 10.1109/CDC.2014.7039601
Pieter Abbeel, Teodor M. Moldovan, Safe Exploration in Markov Decision Processes international conference on machine learning. pp. 1451- 1458 ,(2012)
Yishay Mansour, Satinder P. Singh, Richard S Sutton, David A. McAllester, Policy Gradient Methods for Reinforcement Learning with Function Approximation neural information processing systems. ,vol. 12, pp. 1057- 1063 ,(1999)
Andrew G. Barto, Theodore J. Perkins, Lyapunov design for safe reinforcement learning Journal of Machine Learning Research. ,vol. 3, pp. 803- 832 ,(2003) , 10.5555/944919.944955
Guy Katz, Clark Barrett, David L. Dill, Kyle Julian, Mykel J. Kochenderfer, Reluplex: An Efficient SMT Solver for Verifying Deep Neural Networks computer aided verification. pp. 97- 117 ,(2017) , 10.1007/978-3-319-63387-9_5
Felix Berkenkamp, Andreas Krause, Matteo Turchetta, Angela P. Schoellig, Safe Model-based Reinforcement Learning with Stability Guarantees neural information processing systems. ,vol. 30, pp. 908- 918 ,(2017)
Krishnamurthy Dvijotham, Yuval Tassa, Cosmin Paduraru, Todd Hester, Gal Dalal, Matej Vecerik, Safe Exploration in Continuous Action Spaces arXiv: Artificial Intelligence. ,(2018)