A Reinforcement Learning Algorithm Based on Policy Iteration for Average Reward: Empirical Results with Yield Management and Convergence Analysis

91 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Safe Policy Improvement with Baseline Bootstrapping (2017, background): “Then, the problem is cast without any constraint in a well deﬁned semi-MDP, and Policy Iteration is known to converge in semi-MDPs to the policy optimizing the value function (Gosavi, 2004).”
Finite time bounds for sampling based fitted value iteration (2005, background): “, 1997), dynamic channel allocation (Singh and Bertsekas, 1997), or airline seat allocation (Gosavi, 2004).”
Infinite-Horizon Model Predictive Control for Periodic Tasks with Contacts (2011, methodology): “The optimization criterion of infinite-horizon average-cost has been studied in the past two decades [20, 10, 13, 21]; it was used for policy search [3], and successfully applied to gait optimization [23].”
Finite-Time Bounds for Fitted Value Iteration (2008, background): “…such as job-shop scheduling (Zhang and Dietterich, 1995), backgammon (Tesauro, 1995), elevator control (Crites and Barto, 1995), machine maintenance (Mahadevan et al., 1997), dynamic channel allocation (Singh and Munos and Szepesvári Bertsekas, 1997), or airline seat allocation (Gosavi, 2004).”
Reinforcement Learning: A Tutorial Survey and Recent Advances (2009, unknown): “A well-known " revenue management problem " can be set up as an average-reward SMDP (Gosavi, 2004b; Gosavi et al., 2002).”