Stability-based Generalization Analysis of Randomized Coordinate Descent for Pairwise Learning
This work addresses a theoretical gap for researchers in optimization and machine learning, but it is incremental as it extends existing stability analysis to a specific method and framework.
The paper tackles the lack of theoretical generalization analysis for randomized coordinate descent (RCD) in pairwise learning, such as ranking and metric learning, by deriving generalization bounds based on on-average argument stability for convex and strongly convex objectives, achieving an optimistic excess risk bound of O(1/n) with sample size n.
Pairwise learning includes various machine learning tasks, with ranking and metric learning serving as the primary representatives. While randomized coordinate descent (RCD) is popular in various learning problems, there is much less theoretical analysis on the generalization behavior of models trained by RCD, especially under the pairwise learning framework. In this paper, we consider the generalization of RCD for pairwise learning. We measure the on-average argument stability for both convex and strongly convex objective functions, based on which we develop generalization bounds in expectation. The early-stopping strategy is adopted to quantify the balance between estimation and optimization. Our analysis further incorporates the low-noise setting into the excess risk bound to achieve the optimistic bound as $O(1/n)$, where $n$ is the sample size.