Finite-Sum Coupled Compositional Stochastic Optimization: Theory and Applications
This work addresses optimization challenges in machine learning for compositional measures, offering incremental improvements in algorithm efficiency for specific applications.
The paper tackles finite-sum coupled compositional optimization (FCCO) problems, which arise in machine learning for objectives like average precision and listwise ranking losses, by providing a comprehensive convergence analysis of a stochastic algorithm that achieves improved oracle complexity with parallel speed-up using moving-average estimators and mini-batching.
This paper studies stochastic optimization for a sum of compositional functions, where the inner-level function of each summand is coupled with the corresponding summation index. We refer to this family of problems as finite-sum coupled compositional optimization (FCCO). It has broad applications in machine learning for optimizing non-convex or convex compositional measures/objectives such as average precision (AP), p-norm push, listwise ranking losses, neighborhood component analysis (NCA), deep survival analysis, deep latent variable models, etc., which deserves finer analysis. Yet, existing algorithms and analyses are restricted in one or other aspects. The contribution of this paper is to provide a comprehensive convergence analysis of a simple stochastic algorithm for both non-convex and convex objectives. Our key result is the improved oracle complexity with the parallel speed-up by using the moving-average based estimator with mini-batching. Our theoretical analysis also exhibits new insights for improving the practical implementation by sampling the batches of equal size for the outer and inner levels. Numerical experiments on AP maximization, NCA, and p-norm push corroborate some aspects of the theory.