Robust Solutions to Markov Decision Problems with Uncertain Transition Matrices

78 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.

Robustness in Markov Decision Problems with Uncertain Transition Matrices (2003, background): “Repeat until t = 0: (a) For every statei ∈ X and actiona ∈ A, compute, using the bisection algorithm given in [2], a valuêσ i such that σ̂ i − 2/N ≤ σPa i (v̂t) ≤ σ̂ i .”
Learning decisions: robustness, uncertainty, and approximation (2004, background): “In very recent work [Nilim and el Ghaoui, 2004], a duality result given first (to the best of our knowledge) in [Fleming and Hernandez-Hernandez, 1997] is generalized to make the solution to the inner optimization of Robust Value Iteration very fast for relative entropy balls.”