English

Cautious Weight Decay

Machine Learning 2026-02-25 v2 Optimization and Control Machine Learning

Abstract

We introduce Cautious Weight Decay (CWD), a one-line, optimizer-agnostic modification that applies weight decay only to parameter coordinates whose signs align with the optimizer update. Unlike standard decoupled decay, which implicitly optimizes a regularized or constrained objective, CWD preserves the original loss and admits a bilevel interpretation: it induces sliding-mode behavior upon reaching the stationary manifold, allowing it to search for locally Pareto-optimal stationary points of the unmodified objective. In practice, CWD is a drop-in change for optimizers such as AdamW, Lion, and Muon, requiring no new hyperparameters or additional tuning. For language model pre-training and ImageNet classification, CWD consistently improves final loss and accuracy at million- to billion-parameter scales.

Keywords

Cite

@article{arxiv.2510.12402,
  title  = {Cautious Weight Decay},
  author = {Lizhang Chen and Jonathan Li and Kaizhao Liang and Baiyu Su and Cong Xie and Nuo Wang Pierse and Chen Liang and Ni Lao and Qiang Liu},
  journal= {arXiv preprint arXiv:2510.12402},
  year   = {2026}
}
R2 v1 2026-07-01T06:36:14.184Z