Competitive on-line learning with a convex loss function

dc.creatorVovk, Vladimir
dc.date2005-06-11
dc.date2005-09-02
dc.date.accessioned2026-07-07T03:23:07Z
dc.date.available2026-07-07T03:23:07Z
dc.descriptionWe consider the problem of sequential decision making under uncertainty in which the loss caused by a decision depends on the following binary observation. In competitive on-line learning, the goal is to design decision algorithms that are almost as good as the best decision rules in a wide benchmark class, without making any assumptions about the way the observations are generated. However, standard algorithms in this area can only deal with finite-dimensional (often countable) benchmark classes. In this paper we give similar results for decision rules ranging over an arbitrary reproducing kernel Hilbert space. For example, it is shown that for a wide class of loss functions (including the standard square, absolute, and log loss functions) the average loss of the master algorithm, over the first $N$ observations, does not exceed the average loss of the best decision rule with a bounded norm plus $O(N^{-1/2})$. Our proof technique is very different from the standard ones and is based on recent results about defensive forecasting. Given the probabilities produced by a defensive forecasting algorithm, which are known to be well calibrated and to have good resolution in the long run, we use the expected loss minimization principle to find a suitable decision.
dc.description26 pages
dc.identifierhttps://arxiv.org/abs/cs/0506041
dc.identifierhttp://arxiv.org/abs/cs/0506041
dc.identifier.urihttp://salesiana.dossiersoluciones.com/handle/123456789/32817
dc.subjectMachine Learning
dc.subjectArtificial Intelligence
dc.subjectI.2.6; I.5.1
dc.titleCompetitive on-line learning with a convex loss function
dc.typetext

Files

Collections