Computationally efficient adaptive control algorithms for Markov chains

By A. Jalali, M. Ferguson. 1989

How researchers see this paper

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

Citations in the corpus

Learning to Act Using Real-Time Dynamic Programming (1995, methodology): “Lin [46,47] also discusses methods closely related to Adaptive RTDP In the engineering literature, Jalali and Ferguson [32] describe an algorithm that is similar to Adaptive RTDP, although they focus on Markovian decision problems in which performance is measured by the average cost per time step in”
Average Reward Reinforcement Learning: Foundations, Algorithms, and Empirical Results (2005, methodology): “However, some additional assumptions are also necessary to guarantee convergence, such as that the MDP is identi able by an MLE-estimator (for details of the proof, see [16]).”
Scaling Up Average Reward Reinforcement Learning by Approximating the Domain Models and the Value Function (1996, methodology): “Jalali, A., & Ferguson, M. (1989).”
Average reward reinforcement learning: Foundations, algorithms, and empirical results (2004, methodology): “We focus on one particular algorithm by Jalali and Ferguson ( Jalali & Ferguson, 1989 ).”
An Average-Reward Reinforcement Learning Algorithm for Computing Bias-Optimal Policies (1996, methodology): “The algorithm estimates the transition probabilities from online experience, similar to to Jalali and Ferguson's B algorithm [8], and Tadepalli and Ok's H algorithm [20].”
Reinforcement Learning is Direct Adaptive Optimal Control (1992, methodology): “Q-learning andother reinforcementlearning methods aremostclosely related tobutweredeveloped independently of-theadaptive Markovcontrol method ofJalali andFerguon [ 16 ], whichtheycallasynchronous transient programming." Asynchronous DPmethods described byBertsekas andTsitsiklis [9] perform local upd”