Accelerated Over-Relaxation Heavy-Ball Method: Achieving Global Accelerated Convergence with Broad Generalization

Abstract

The heavy-ball momentum method accelerates gradient descent with a momentum term but lacks accelerated convergence for general smooth strongly convex problems. This work introduces the Accelerated Over-Relaxation Heavy-Ball (AOR-HB) method, the first variant with provable global and accelerated convergence for such problems. AOR-HB closes a long-standing theoretical gap, extends to composite convex optimization and min-max problems, and achieves optimal complexity bounds. It offers three key advantages: (1) broad generalization ability, (2) potential to reshape acceleration techniques, and (3) conceptual clarity and elegance compared to existing methods.

Cite

Text

Wei and Chen. "Accelerated Over-Relaxation Heavy-Ball Method: Achieving Global Accelerated Convergence with Broad Generalization." International Conference on Learning Representations, 2025.

Markdown

[Wei and Chen. "Accelerated Over-Relaxation Heavy-Ball Method: Achieving Global Accelerated Convergence with Broad Generalization." International Conference on Learning Representations, 2025.](https://mlanthology.org/iclr/2025/wei2025iclr-accelerated/)

BibTeX

@inproceedings{wei2025iclr-accelerated,
  title     = {{Accelerated Over-Relaxation Heavy-Ball Method: Achieving Global Accelerated Convergence with Broad Generalization}},
  author    = {Wei, Jingrong and Chen, Long},
  booktitle = {International Conference on Learning Representations},
  year      = {2025},
  url       = {https://mlanthology.org/iclr/2025/wei2025iclr-accelerated/}
}