Ultraconservative Online Algorithms for Multiclass Problems

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 ultracon-servativeness. 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.

[1]  F ROSENBLATT,et al.  The perceptron: a probabilistic model for information storage and organization in the brain. , 1958, Psychological review.

[2]  Richard O. Duda,et al.  Pattern classification and scene analysis , 1974, A Wiley-Interscience publication.

[3]  Nick Littlestone.  Learning Quickly When Irrelevant Attributes Abound: A New Linear-Threshold Algorithm (Extended Abstract) , 1987, FOCS.

[4]  J. Ross Quinlan,et al.  C4.5: Programs for Machine Learning , 1992 .

[5]  Manfred K. Warmuth,et al.  Additive versus exponentiated gradient updates for linear prediction , 1995, STOC '95.

[6]  Manfred K. Warmuth,et al.  On Weak Learning , 1995, J. Comput. Syst. Sci..

[7]  Thomas G. Dietterich,et al.  Solving Multiclass Learning Problems via Error-Correcting Output Codes , 1994, J. Artif. Intell. Res..

[8]  Dale Schuurmans,et al.  General Convergence Results for Linear Discriminant Updates , 1997, COLT '97.

[9]  Manfred K. Warmuth,et al.  Exponentiated Gradient Versus Gradient Descent for Linear Predictors , 1997, Inf. Comput..

[10]  Yoav Freund,et al.  Large Margin Classification Using the Perceptron Algorithm , 1998, COLT' 98.

[11]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .

[12]  Nello Cristianini,et al.  The Kernel-Adatron Algorithm: A Fast and Simple Learning Procedure for Support Vector Machines , 1998, ICML.

[13]  Nello Cristianini,et al.  The Kernel-Adatron : A fast and simple learning procedure for support vector machines , 1998, ICML 1998.

[14]  Chris Mesterharm.  A Multi-class Linear Learning Algorithm Related to Winnow , 1999, NIPS.

[15]  John C. Platt,et al.  Fast training of support vector machines using sequential minimal optimization, advances in kernel methods , 1999 .

[16]  Yoram Singer,et al.  Reducing Multiclass to Binary: A Unifying Approach for Margin Classifiers , 2000, J. Mach. Learn. Res..

[17]  Claudio Gentile,et al.  A New Approximate Maximal Margin Classification Algorithm , 2002, J. Mach. Learn. Res..

[18]  Koby Crammer,et al.  On the Algorithmic Implementation of Multiclass Kernel-based Vector Machines , 2002, J. Mach. Learn. Res..

[19]  Yi Li,et al.  The Relaxed Online Maximum Margin Algorithm , 1999, Machine Learning.

[20]  Koby Crammer,et al.  On the Learnability and Design of Output Codes for Multiclass Problems , 2002, Machine Learning.

[21]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.