Efficient Algorithms for Combinatorial Online Prediction

Bibliographic Information

Published
2013
Resource Type
journal article
DOI
  • 10.1007/978-3-642-40935-6_3
Publisher
Springer Berlin Heidelberg

Search this article

Description

We study online linear optimization problems over concept classes which are defined in some combinatorial ways. Typically, those concept classes contain finite but exponentially many concepts and hence the complexity issue arises. In this paper, we survey some recent results on universal and efficient implementations of low-regret algorithmic frameworks such as Follow the Regularized Leader FTRL and Follow the Perturbed Leader FPL.

Journal

References(16)*help

See more

Related Projects

See more

Details 詳細情報について

Report a problem

Back to top