Semistochastic Quadratic Bound Methods

TL;DR

This paper introduces semistochastic quadratic bound methods for efficient maximum likelihood inference in models involving partition functions, demonstrating their convergence properties and superior performance over existing techniques.

Contribution

It develops semistochastic quadratic bound algorithms with proven convergence and stability enhancements, advancing optimization methods for partition function-based models.

Findings

Proven global convergence to stationary points.

Achieved linear convergence rate under certain conditions.

Demonstrated superior performance on benchmark datasets.

Abstract

Partition functions arise in a variety of settings, including conditional random fields, logistic regression, and latent gaussian models. In this paper, we consider semistochastic quadratic bound (SQB) methods for maximum likelihood inference based on partition function optimization. Batch methods based on the quadratic bound were recently proposed for this class of problems, and performed favorably in comparison to state-of-the-art techniques. Semistochastic methods fall in between batch algorithms, which use all the data, and stochastic gradient type methods, which use small random selections at each iteration. We build semistochastic quadratic bound-based methods, and prove both global convergence (to a stationary point) under very weak assumptions, and linear convergence rate under stronger assumptions on the objective. To make the proposed methods faster and more stable, we consider inexact subproblem minimization and batch-size selection schemes. The efficacy of SQB methods is demonstrated via comparison with several state-of-the-art techniques on commonly used datasets.