WORST-CASE QUADRATIC LOSS BOUNDS FOR ON-LINE PREDICTION OF LINEAR FUNCTIONS BY GRADIENT DESCENT

In this paper we study the performance of gradient descent when applied to the problem of on-line linear prediction in arbitrary inner product spaces. We show worst-case bounds on the sum of the squared prediction errors under various assumptions concerning the amount of a priori information about the sequence to predict. The algorithms we use are variants and extensions of on-line gradient descent. Whereas our algorithms always predict using linear functions as hypotheses, none of our results requires the data to be linearly related. In fact, the bounds proved on the total prediction loss are typically expressed as a function of the total loss of the best fixed linear predictor with bounded norm. All the upper bounds are tight to within constants. Matching lower bounds are provided in some cases. Finally, we apply our results to the problem of on-line prediction for classes of smooth functions.

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

[2]  Charles R. Johnson,et al.  Matrix analysis , 1985, Statistical Inference for Engineers and Data Scientists.

[3]  Michael I. Jordan,et al.  Advances in Neural Information Processing Systems 30 , 1995 .

[4]  Vladimir Vovk,et al.  Aggregating strategies , 1990, COLT '90.

[5]  M. Zwaan.  An introduction to hilbert space , 1990 .

[6]  Philip M. Long,et al.  On-line learning of linear functions , 1991, STOC '91.

[7]  Neri Merhav,et al.  Universal sequential learning and decision from individual data sequences , 1992, COLT '92.

[8]  Philip M. Long,et al.  The learning complexity of smooth functions of a single variable , 1992, COLT '92.

[9]  Neri Merhav,et al.  Universal prediction of individual sequences , 1992, IEEE Trans. Inf. Theory.

[10]  David Haussler,et al.  How to use expert advice , 1993, STOC.

[11]  Manfred K. Warmuth,et al.  The Weighted Majority Algorithm , 1994, Inf. Comput..