English

Understanding Incremental Learning with Closed-form Solution to Gradient Flow on Overparamerterized Matrix Factorization

Machine Learning 2025-08-29 v1 Systems and Control Systems and Control

Abstract

Many theoretical studies on neural networks attribute their excellent empirical performance to the implicit bias or regularization induced by first-order optimization algorithms when training networks under certain initialization assumptions. One example is the incremental learning phenomenon in gradient flow (GF) on an overparamerterized matrix factorization problem with small initialization: GF learns a target matrix by sequentially learning its singular values in decreasing order of magnitude over time. In this paper, we develop a quantitative understanding of this incremental learning behavior for GF on the symmetric matrix factorization problem, using its closed-form solution obtained by solving a Riccati-like matrix differential equation. We show that incremental learning emerges from some time-scale separation among dynamics corresponding to learning different components in the target matrix. By decreasing the initialization scale, these time-scale separations become more prominent, allowing one to find low-rank approximations of the target matrix. Lastly, we discuss the possible avenues for extending this analysis to asymmetric matrix factorization problems.

Keywords

Cite

@article{arxiv.2508.20344,
  title  = {Understanding Incremental Learning with Closed-form Solution to Gradient Flow on Overparamerterized Matrix Factorization},
  author = {Hancheng Min and René Vidal},
  journal= {arXiv preprint arXiv:2508.20344},
  year   = {2025}
}

Comments

Accepted to CDC 2025

R2 v1 2026-07-01T05:09:28.709Z