资源分类
》 资源 》论文》New Adaptive Algorithms for Online Classification

New Adaptive Algorithms for Online Classification

2020-01-06 | |  95 |   37 |   0

Abstract

We propose a general framework to online learning for classification problems with time-varying potential functions in the adversarial setting. This framework allows to design and prove relative mistake bounds for any generic loss function. The mistake bounds can be specialized for the hinge loss, allowing to recover and improve the bounds of known online classification algorithms. By optimizing the general bound we derive a new online classification algorithm, called NAROW, that hybridly uses adaptiveand fixedsecond order information. We analyze the properties of the algorithm and illustrate its performance using synthetic dataset.

上一篇：Fast global convergence of gradient methods for high-dimensional statistical recovery

下一篇：Tight Sample Complexity of Large-Margin Learning

用户评价
全部评价

热门资源

  • The Variational S...

    Unlike traditional images which do not offer in...

  • Learning to Predi...

    Much of model-based reinforcement learning invo...

  • Stratified Strate...

    In this paper we introduce Stratified Strategy ...

  • A Mathematical Mo...

    Direct democracy, where each voter casts one vo...

  • Rating-Boosted La...

    The performance of a recommendation system reli...