Near-Optimal Algorithms for Online Matrix Prediction

Abstract

In several online prediction problems of recent interest the comparison class is composed of matrices with bounded entries. For example, in the online max-cut problem, the comparison class is matrices which represent cuts of a given graph and in online gambling the comparison class is matrices which represent permutations over n teams. Another important example is online collaborative filtering in which a widely used comparison class is the set of matrices with a small trace norm. In this paper we isolate a property of matrices, which we call (beta,tau)-decomposability, and derive an efficient online learning algorithm, that enjoys a regret bound of O*(sqrt(beta tau T)) for all problems in which the comparison class is composed of (beta,tau)-decomposable matrices. By analyzing the decomposability of cut matrices, triangular matrices, and low trace-norm matrices, we derive near optimal regret bounds for online max-cut, online gambling, and online collaborative filtering. In particular, this resolves (in the affirmative) an open problem posed by Abernethy (2010); Kleinberg et al (2010). Finally, we derive lower bounds for the three problems and show that our upper bounds are optimal up to logarithmic factors. In particular, our lower bound for the online collaborative filtering problem resolves another open problem posed by Shamir and Srebro (2011).

25 pages

Cited by in corpus (18)

• Novel methods for multilinear data completion and de-noising based on tensor-SVD
• More data speeds up training time in learning halfspaces over sparse vectors
• A Tour of Reinforcement Learning: The View from Continuous Control
• Classification with Low Rank and Missing Data
• Novel Factorization Strategies for Higher Order Tensors: Implications for Compression and Recovery of Multi-linear Data
• Online Similarity Prediction of Networked Data from Known and Unknown Graphs
• Exponentiated Gradient Meets Gradient Descent
• Matrix reconstruction with the local max norm
• Maximum entropy low-rank matrix recovery
• Label optimal regret bounds for online local learning
• Efficient Online Learning of Optimal Rankings: Dimensionality Reduction via Gradient Descent
• On the Regret Minimization of Nonconvex Online Gradient Ascent for Online PCA
• An algorithm for online tensor prediction
• Stochastic Multiplicative Weights Updates in Zero-Sum Games
• Adaptive and Efficient Algorithms for Tracking the Best Expert
• Online Matrix Completion with Side Information
• Adversarial Online Learning with Changing Action Sets: Efficient Algorithms with Approximate Regret Bounds
• A generalised log-determinant regularizer for online semi-definite programming and its applications