On Sequences with Non-Learnable Subsequences

dc.creatorV'yugin, Vladimir V.
dc.date2008-06-26
dc.date.accessioned2026-07-07T09:46:56Z
dc.date.available2026-07-07T09:46:56Z
dc.descriptionThe remarkable results of Foster and Vohra was a starting point for a series of papers which show that any sequence of outcomes can be learned (with no prior knowledge) using some universal randomized forecasting algorithm and forecast-dependent checking rules. We show that for the class of all computationally efficient outcome-forecast-based checking rules, this property is violated. Moreover, we present a probabilistic algorithm generating with probability close to one a sequence with a subsequence which simultaneously miscalibrates all partially weakly computable randomized forecasting algorithms. %subsequences non-learnable by each randomized algorithm. According to the Dawid's prequential framework we consider partial recursive randomized algorithms.
dc.identifierhttps://arxiv.org/abs/0806.4341
dc.identifierhttp://arxiv.org/abs/0806.4341
dc.identifierLNCS 5010, pp. 302-313, 2008
dc.identifier.urihttp://salesiana.dossiersoluciones.com/handle/123456789/163700
dc.subjectArtificial Intelligence
dc.subjectMachine Learning
dc.subjectF.4.1; I.2.6
dc.titleOn Sequences with Non-Learnable Subsequences
dc.typetext

Files

Collections