paper-with-me

Papers

Open Problem: Is AdamW Effective Under Heavy-Tailed Noise?

2026-06-22 · Dingzhi Yu, Hongyi Tao, Yuanyu Wan, Luo Luo, Lijun Zhang arxiv

AdamW is the de facto optimizer for training large language models (LLMs), yet the theory behind it still lives mostly in finite-variance regimes. This is increasingly unsatisfying, as empirical evidence indicates that stochastic gradient noise in LLM pretraining is typically heavy-tailed. Recent work shows that sign-based optimizers such as Lion and Muon achieve sharp heavy-tailed rates, and that AdaGrad can also converge under heavy-tailed noise. However, no rigorous convergence theory for AdamW has yet been established in this regime. Can AdamW converge under the same heavy-tailed assumptions, or does its second-moment accumulator create a genuine obstruction? We formulate this as an open problem, prove a positive weighted-metric benchmark, and give a corridor lower-bound mechanism showing how denominator memory can hide large gradients.

📄 PDF Abstract BibTeX arXiv:2606.23676

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Muon in Vision Transformers: Optimizer-Recipe Interactions and Gradient Spectra

2026-05-23 · Ben S. Southworth, Shuai Jiang, Daniel McBride, Eric C. Cyr 외 arxiv

Muon is a recently developed matrix-aware optimizer that has shown strong results in transformer training, but its behavior in vision transformers (ViTs) is not yet well understood. We study Muon for ViT training, largel…

Image SegmentationData Augmentation

Sign-Based Optimizers Are Effective Under Heavy-Tailed Noise

2026-02-07 · Dingzhi Yu, Hongyi Tao, Yuanyu Wan, Luo Luo 외 arxiv

While adaptive gradient methods are the workhorse of modern machine learning, sign-based optimization algorithms such as Lion and Muon have recently demonstrated superior empirical performance over AdamW in training larg…

FedAdamW: A Communication-Efficient Optimizer with Convergence and Generalization Guarantees for Federated Large Models

2025-10-31 · Junkang Liu, Fanhua Shang, Hongying Liu, Yuxuan Tian 외 arxiv

AdamW has become one of the most effective optimizers for training large-scale models. We have also observed its effectiveness in the context of federated learning (FL). However, directly applying AdamW in federated lear…

Federated Learning

LionMuon: Alternating Spectral and Sign Descent for Efficient Training

2026-05-19 · Arman Bolatov, Artem Riabinin, Nikita Kornilov, Andrey Veprikov 외 arxiv

In large-scale optimization, the cheapness and effectiveness of update steps are the most crucial factors for a successful optimizer. Sign-based optimizers like Lion or Signum produce cheap per-step updates, whereas Muon…

Adaptive Memory Momentum via a Model-Based Framework for Deep Learning Optimization

2025-10-06 · Kristi Topollai, Anna Choromanska arxiv

The vast majority of modern deep learning models are trained with momentum-based first-order optimizers. The momentum term governs the optimizer's memory by determining how much each past gradient contributes to the curr…