×

Tracking the best of many experts. (English) Zbl 1137.68540

Auer, Peter (ed.) et al., Learning theory. 18th annual conference on learning theory, COLT 2005, Bertinoro, Italy, June 27–30, 2005. Proceedings. Berlin: Springer (ISBN 3-540-26556-2/pbk). Lecture Notes in Computer Science 3559. Lecture Notes in Artificial Intelligence, 204-216 (2005).
Summary: An algorithm is presented for online prediction that allows to track the best expert efficiently even if the number of experts is exponentially large, provided that the set of experts has a certain structure allowing efficient implementations of the exponentially weighted average predictor. As an example we work out the case where each expert is represented by a path in a directed graph and the loss of each expert is the sum of the weights over the edges in the path.
For the entire collection see [Zbl 1076.68003].

MSC:

68T05 Learning and adaptive systems in artificial intelligence
PDFBibTeX XMLCite
Full Text: DOI