Ultraconservative Online Algorithms for Multiclass Problems

DMCA

Ultraconservative Online Algorithms for Multiclass Problems (2001)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Koby Crammer , Yoram Singer

Venue:	Journal of Machine Learning Research
Citations:	320 - 21 self

BibTeX

@ARTICLE{Crammer01ultraconservativeonline,
    author = {Koby Crammer and Yoram Singer},
    title = {Ultraconservative Online Algorithms for Multiclass Problems},
    journal = {Journal of Machine Learning Research},
    year = {2001},
    volume = {3},
    pages = {2003}
}

OpenURL

Abstract

In this paper we study online classification algorithms for multiclass problems in the mistake bound model. The hypotheses we use maintain one prototype vector per class. Given an input instance, a multiclass hypothesis computes a similarity-score between each prototype and the input instance and then sets the predicted label to be the index of the prototype achieving the highest similarity. To design and analyze the learning algorithms in this paper we introduce the notion of ultraconservativeness. Ultraconservative algorithms are algorithms that update only the prototypes attaining similarity-scores which are higher than the score of the correct label's prototype. We start by describing a family of additive ultraconservative algorithms where each algorithm in the family updates its prototypes by finding a feasible solution for a set of linear constraints that depend on the instantaneous similarity-scores. We then discuss a specific online algorithm that seeks a set of prototypes which have a small norm. The resulting algorithm, which we term MIRA (for Margin Infused Relaxed Algorithm) is ultraconservative as well. We derive mistake bounds for all the algorithms and provide further analysis of MIRA using a generalized notion of the margin for multiclass problems.

Keyphrases

multiclass problem ultraconservative online algorithm input instance relaxed algorithm predicted label prototype vector term mira additive ultraconservative algorithm generalized notion linear constraint instantaneous similarity-scores mistake bound model small norm ultraconservative algorithm online classification algorithm feasible solution learning algorithm specific online algorithm multiclass hypothesis correct label

DMCA