Strategies for prediction under imperfect monitoring

dc.creatorLugosi, Gabor
dc.creatorMannor, Shie
dc.creatorStoltz, Gilles
dc.date2007-01-15
dc.date2008-01-07
dc.date.accessioned2026-07-07T08:52:40Z
dc.date.available2026-07-07T08:52:40Z
dc.descriptionWe propose simple randomized strategies for sequential prediction under imperfect monitoring, that is, when the 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. It was Rustichini (1999) 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, these rates are optimal up to logarithmic terms.
dc.descriptionJournal version of a COLT conference paper
dc.identifierhttps://arxiv.org/abs/math/0701419
dc.identifierhttp://arxiv.org/abs/math/0701419
dc.identifierMathematics of Operations Research (2008) à paraître
dc.identifier.urihttp://salesiana.dossiersoluciones.com/handle/123456789/145359
dc.subjectStatistics Theory
dc.subjectMachine Learning
dc.subject91A20, 62L12, 68Q32
dc.titleStrategies for prediction under imperfect monitoring
dc.typetext

Files

Collections