Abstract | ||
---|---|---|
We propose simple randomized strategies for sequential decision (or prediction) under imperfect monitoring, that is, when the decision maker (forecaster) does not have access to the past outcomes but rather to a feedback signal. The proposed strategies are consistent in the sense that they achieve, asymptotically, the best-possible average reward among all fixed actions. It was Rustichini [Rustichini, A. 1999. Minimizing regret: The general case. Games Econom. Behav.29 224--243] who first proved the existence of such consistent predictors. The forecasters presented here offer the first constructive proof of consistency. Moreover, the proposed algorithms are computationally efficient. We also establish upper bounds for the rates of convergence. In the case of deterministic feedback signals, these rates are optimal up to logarithmic terms. |
Year | DOI | Venue |
---|---|---|
2008 | 10.1287/moor.1080.0312 | Clinical Orthopaedics and Related Research |
Keywords | DocType | Volume |
possible average reward,deterministic feedback signal,sequential decision,consistent predictor,regret,general case,feedback signal,proposed algorithm,decision maker,past outcome,minimizing regret,constructive proof,games econom,hannan consistency,imperfect monitoring,proposed strategy,repeated games,deterministic feedback,on-line learning,sequential prediction | Journal | 33 |
Issue | ISSN | Citations |
3 | Mathematics of Operations Research (2008) \`a para\^itre | 10 |
PageRank | References | Authors |
0.98 | 14 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
GáBor Lugosi | 1 | 1092 | 195.02 |
Shie Mannor | 2 | 3340 | 285.45 |
Gilles Stoltz | 3 | 351 | 31.53 |