Thumbnail

History-dependent evaluations in POMDPs

Venel, Xavier; Ziliotto, Bruno (2021), History-dependent evaluations in POMDPs, SIAM Journal on Control and Optimization, 59, 2, p. 1730–1755. 10.1137/20M1332876

View/Open
Type
Article accepté pour publication ou publié
Date
2021
Journal name
SIAM Journal on Control and Optimization
Volume
59
Number
2
Publisher
SIAM - Society for Industrial and Applied Mathematics
Pages
1730–1755
Publication identifier
Metadata
Show full item record
Author(s)
Venel, Xavier cc
Centre d'économie de la Sorbonne [CES]
Ziliotto, Bruno
CEntre de REcherches en MAthématiques de la DEcision [CEREMADE]
Abstract (EN)
We consider POMDPs in which the weight of the stage payoff depends on the past sequence of signals and actions occurring in the infinitely repeated problem. We prove that for all epsilon>0, there exists a strategy that is epsilon-optimal for any sequence of weights satisfying a property that interprets as "the decision-maker is patient enough". This unifies and generalizes several results of the literature, and applies notably to POMDPs with limsup payoffs.
Subjects / Keywords
Markov decision process; partial observation; long-run average payoff