Abstract
We consider a control problem where the decision maker interacts with a standard Markov decision process with the exception that the reward functions vary arbitrarily over time. We extend the notion of Hannan consistency to this setting, showing that, in hindsight, the agent can perform almost as well as every deterministic policy. We present efficient online algorithms in the spirit of reinforcement learning that ensure that the agent’s performance loss, or regret, vanishes over time, provided that the environment is oblivious to the agent’s actions. However, counterexamples indicate that the regret does not vanish if the environment is not oblivious.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Auer, P., Cesa-Bianchi, N., Freund, Y., Schapire, R.E.: The nonstochastic multiarmed bandit problem. SIAM J. Computing 32(1), 48–77 (2002)
Aumann, R.J.: Markets with a continuum of traders. Econometrica 32, 39–50 (1964)
Bertsekas, D.P.: Dynamic programming and optimal control, 2nd edn., vol. 2. Athena Scientific (2001)
Bertsekas, D.P., Tsitsiklis, J.N.: Neuro-dynamic programming. Athena Scientific (1996)
Bobkov, S.G., Tetali, P.: Modified logarithmic Sobolev inequalities in discrete settings. Journal of Theoretical Probability 19(2), 289–336 (2006)
Borkar, V.S., Meyn, S.P.: The O.D.E. method for convergence of stochastic approximation and reinforcement learning. SIAM J. Control and Optimization 38(2), 447–469 (2000)
Brafman, R.I., Tennenholtz, M.: R-max—a general polynomial time algorithm for near-optimal reinforcement learning. Journal of Machine Learning Research 3, 213–231 (2003)
Cesa-Bianchi, N., Lugosi, G.: Prediction, learning, and games. Cambridge University Press, Cambridge (2006)
Crites, R.H., Barto, A.G.: An actor/critic algorithm that is equivalent to Q-learning. In: Advances in Neural Information Processing Systems, pp. 401–408 (1995)
Duffield, N.G., Massey, W.A., Whitt, W.: A nonstationary offered-load model for packet networks. Telecommunication Systems 16(3–4), 271–296 (2001)
Even-Dar, E., Kakade, S., Mansour, Y.: Experts in a Markov decision process. In: NIPS, pp. 401–408 (2004)
Fudenberg, D., Kreps, D.M.: Learning mixed equilibria. Games and Economic Behavior 5(3), 320–367 (1993)
Hannan, J.: Approximation to Bayes risk in repeated play. In: Contributions to the Theory of Games, vol. 3, pp. 97–139. Princeton University Press, Princeton (1957)
Herbster, M., Warmuth, M.K.: Tracking the best expert. Machine Learning 32(2), 151–178 (1998)
Kalai, A., Vempala, S.: Efficient algorithms for online decision problems. Journal of Computer and System Sciences 71(3), 291–307 (2005)
Mannor, S., Shimkin, N.: The empirical Bayes envelope and regret minimization in competitive Markov decision processes. Mathematics of Operations Research 28(2), 327–345 (2003)
Merhav, N., Ordentlich, E., Seroussi, G., Weinberger, M.J.: On sequential strategies for loss functions with memory. IEEE Trans. Inf. Theory 48(7), 1947–1958 (2002)
Shapley, L.: Stochastic games. PNAS 39(10), 1095–1100 (1953)
Yu, J.Y., Mannor, S., Shimkin, N.: Markov decision processes with arbitrarily varying rewards (Preprint, 2008), http://www.cim.mcgill.ca/~jiayuan/mdp.pdf
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2008 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Yu, J.Y., Mannor, S., Shimkin, N. (2008). Markov Decision Processes with Arbitrary Reward Processes. In: Girgin, S., Loth, M., Munos, R., Preux, P., Ryabko, D. (eds) Recent Advances in Reinforcement Learning. EWRL 2008. Lecture Notes in Computer Science(), vol 5323. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-89722-4_21
Download citation
DOI: https://doi.org/10.1007/978-3-540-89722-4_21
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-89721-7
Online ISBN: 978-3-540-89722-4
eBook Packages: Computer ScienceComputer Science (R0)