Sample efficient inductive matrix completion with noise and inexact side information

2026-05-16

AI Summary

This work addresses the problem of sample-efficient inductive matrix completion under noisy observations and imperfect row/column side information. The authors propose a non-convex projected gradient descent algorithm enhanced with spectral initialization, which—under this challenging setting—is the first to achieve a sample complexity that depends only on the effective problem dimension rather than the ambient matrix dimensions. The method guarantees linear convergence and attains minimax-optimal estimation error. Convergence and statistical optimality are established through a regularity condition in the theoretical analysis. Extensive experiments on the MovieLens dataset and synthetic data demonstrate the algorithm’s superior sample efficiency and robustness to noise and inaccuracies in side information.

Abstract

Low-rank matrix completion is a widely studied problem with many variants. Inductive matrix completion (IMC) incorporates row and column side information to significantly narrow the search space. Prior work falls into two regimes: methods that exploit this structure to achieve reduced sample complexity but only in noiseless settings, and methods that handle noise but require sample complexity matching the ambient matrix dimension, forfeiting the sample efficiency that side information should provide. In this paper, we close this gap by studying noisy IMC with a nonconvex projected gradient descent algorithm with spectral initialization. Our main technical contribution is establishing a regularity condition for the IMC loss function that holds at the reduced sample complexity determined by the effective problem size, scaling with the side information dimension a rather than the ambient dimension n. This directly yields linear convergence and an estimation error that both depend only on the effective problem size rather than the ambient matrix dimension. We further extend our analysis to the inexact side information setting, demonstrating that the reduced sample complexity is maintained and the estimation error is order-optimal with respect to the inexactness of the side information. Extensive simulations and real-world experiments on the MovieLens dataset validate our theoretical findings.

Problem

• inductive matrix completion
• sample efficiency
• noise
• inexact side information
• low-rank matrix completion

Innovation

• inductive matrix completion
• sample efficiency
• nonconvex optimization
• spectral initialization
• side information