A primal-dual perspective of online learning algorithms

Abstract We describe a novel framework for the design and analysis of online learning algorithms based on the notion of duality in constrained optimization. We cast a sub-family of universal online bounds as an optimization problem. Using the weak duality theorem we reduce the process of online learning to the task of incrementally increasing the dual objective function. The amount by which the dual increases serves as a new and natural notion of progress for analyzing online learning algorithms. We are thus able to tie the primal objective value and the number of prediction mistakes using the increase in the dual.

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

[2]  James Hannan,et al.  4. APPROXIMATION TO RAYES RISK IN REPEATED PLAY , 1958 .

[3]  Albert B Novikoff,et al.  ON CONVERGENCE PROOFS FOR PERCEPTRONS , 1963 .

[4]  L. Bregman The relaxation method of finding the common point of convex sets and its application to the solution of problems in convex programming , 1967 .

[5]  丸山 徹 Convex Analysisの二,三の進展について , 1977 .

[6]  N. Littlestone Learning Quickly When Irrelevant Attributes Abound: A New Linear-Threshold Algorithm , 1987, 28th Annual Symposium on Foundations of Computer Science (sfcs 1987).

[7]  N. Littlestone Mistake bounds and logarithmic linear-threshold learning algorithms , 1990 .

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

[9]  Y. Censor,et al.  Parallel Optimization: Theory, Algorithms, and Applications , 1997 .

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

[11]  Claudio Gentile,et al.  The Robustness of the p-Norm Algorithms , 1999, COLT '99.

[12]  Manfred K. Warmuth,et al.  Relative loss bounds for single neurons , 1999, IEEE Trans. Neural Networks.

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

[14]  V. Vovk Competitive On‐line Statistics , 2001 .

[15]  Koby Crammer,et al.  Online Passive-Aggressive Algorithms , 2003, J. Mach. Learn. Res..

[16]  Alexander J. Smola,et al.  Online learning with kernels , 2001, IEEE Transactions on Signal Processing.

[17]  Manfred K. Warmuth,et al.  Relative Loss Bounds for Multidimensional Regression Problems , 1997, Machine Learning.

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

[19]  Manfred K. Warmuth,et al.  Relative Loss Bounds for On-Line Density Estimation with the Exponential Family of Distributions , 1999, Machine Learning.

[20]  Claudio Gentile,et al.  On the generalization ability of on-line learning algorithms , 2001, IEEE Transactions on Information Theory.

[21]  Amos Storkey,et al.  Advances in Neural Information Processing Systems 20 , 2007 .

[22]  Claudio Gentile,et al.  A Second-Order Perceptron Algorithm , 2002, SIAM J. Comput..

[23]  Yoram Singer,et al.  The Forgetron: A Kernel-Based Perceptron on a Fixed Budget , 2005, NIPS.

[24]  Stephen P. Boyd,et al.  Convex Optimization , 2004, Algorithms and Theory of Computation Handbook.