The effect of representation and knowledge on goal-directed exploration with reinforcement-learning algorithms

Machine Learning

Volumn 22, Issue 1-3, 1996, Pages 227-250

  Koenig, Sven ^a   Simmons, Reid G ^a  

^a Carnegie Mellon University   (United States)

Author keywords

Action models; Action penalty representation; Admissible and consistent heuristics; Complexity; Goal reward representation; Goal directed exploration; On line reinforcement learning; Prior knowledge; Q hat learning; Q learning; Reward structure

Indexed keywords

COMPUTATIONAL COMPLEXITY; HEURISTIC METHODS; KNOWLEDGE REPRESENTATION; LEARNING ALGORITHMS; MATHEMATICAL MODELS; ONLINE SYSTEMS; POLYNOMIALS; STATE SPACE METHODS; TOPOLOGY;

ACTION MODELS; GOAL DIRECTED EXPLORATION; GOAL REWARD REPRESENTATION; REINFORCEMENT LEARNING;

LEARNING SYSTEMS;

EID: 0029751419     PISSN: 08856125     EISSN: None     Source Type: Journal    
DOI: 10.1007/BF00114729     Document Type: Article

References (22)

1. Barto, A. G., S.J. Bradtke, and S. P. Singh (1995) Learning to act using real-time dynamic programming. Artificial Intelligence, 73(1):81-138.
2. Barto,A. G., R.S. Sutton, and C.J. Watkins. (1989) Learning and sequential decision making. Technical Report 89-95. Department of Computer Science, University of Massachusetts at Amherst.
3. Bellman, R. (1917) Dynamic Programming. Princeton University Press, Princeton (New Jersey)
4. Boddy,M. and T. Dean. (1989). Solving time-dependent planning problems. In Proceeding of the IJCAI, pages 970-984.
5. Goodwill, R. (1994) Reasoning about when to start acting. In Proceedings of the International Conference on Artificial Intelligence Planning Systems, pages 86-91.
6. Heger,M (1996) The loss from imperfect value functions in expectation based and minimax-based tasks. Machine Learning, pages 197-225.
7. Heger,M. (1994) Consideration of risk in reinforcement learning. In Proceedings of the International Conference on Machine Learning, pages 105-111.
8. Ishida,T (1992). Moving target search with intelligence. In Proceedings of the AAAI, pages 525-532.
9. Ishida, T. and R.E. Korf. (1991). Moving target search. In Proceedings of the IJCAI, pages 204-210.
10. Kaelbling., I. P. (1990). Learning in Entbedded Systems. MIT Press, Cambridge (Massachusetts).
11. Koenig, S (1991) Optimal probabilistic and decision-theoretic planning using Markovian decision theory. Master's thesis, Computer Science Department, University of California at Berkeley
12. (Available as Technical Report UCB/CSD 92/685).
13. Koenig,S. and R.G. Simmons (1992) Complexity analysis of real-time reinforcement learning applied to finding shortest paths in deterministic domains. Technical Report CMU-CS-93-106, School of Computer Science,Carnegie Mellon University.
14. Koenig,S. and R.G. Simmons. (1993). Complexity analysis of real time reinforcement learning. In Proceedings of the AAAI, pages 99-105.
15. Koenig,S. and R.G. Simmons. (1994). How to make reactive planners risk-sensitive. In Proceedings of the International Conference on Artificial Intelligence Planning Systems. page 293-298.
16. Koenig,S. and R.G. Simmons. (1995a) The effect of representation and knowledge on goal-directed exploration with reinforcement learning algorithms the proofs Technical Report CMU-CS-45-177, School of Computer Science, Carnegie Mellon University.
17. Koenig,S and R. G. Simmons (1995b). Real time search in non-deterministic domains. In Proceedings of the IJCAI, pages 1660-1667.
18. Korf, R. F. (1990). Real-time beuristic search. Artificial Intelligence. 42(2-3):189-211.
19. Lin, L. J. (1992) Self-improving reactive agents based on reinforcement learning, planning and teaching. Machine Learning, 8:293-321.
20. Matarić,M. (1994). Interaction and Intelligent Behavior. PhD thesis, Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology
21. Moore, A.W. and C.G. Atkeson (1993a). The parti-game algorithm for variable resolution reinforcement learning in multidimensional state-spaces. In Proceeding of the NIPS.
22. Moore, A.W. Lind C.G. Atkeson. (1993b). Prioritized sweeping: Reinforcement learning with less data and less time. Machine Learning, 13:103-130.