paper-with-me

홈 › Papers

Distributed Sign Momentum with Local Steps for Training Transformers

2024-11-26 · Shuhua Yu, Ding Zhou, Cong Xie, An Xu, Zhi Zhang, Xin Liu, Soummya Kar

Pre-training Transformer models is resource-intensive, and recent studies have shown that sign momentum is an efficient technique for training large-scale deep learning models, particularly Transformers. However, its application in distributed training or federated learning remains underexplored. This paper investigates a novel communication-efficient distributed sign momentum method with local updates. Our proposed method allows for a broad class of base optimizers for local updates, and uses sign momentum in global updates, where momentum is generated from differences accumulated during local steps. We evaluate our method on the pre-training of various GPT-2 models, and the empirical results show significant improvement compared to other distributed methods with local updates. Furthermore, by approximating the sign operator with a randomized version that acts as a continuous analog in expectation, we present an $O(1/\sqrt{T})$ convergence for one instance of the proposed method for nonconvex smooth functions.

📄 PDF Abstract BibTeX arXiv:2411.17866

Code (1)

shuhuayu/dist-sign-momentum 공식 구현 pytorch

Tasks

Federated Learning

Methods 이 논문이 사용한 방법론

Refunds@Expedia|||How do I get a full refund from Expedia? “How do I get a full refund from Expedia? How do I get a full refund from Expedia? – Call ☎️ +1-(888) 829 (0881) or +1-805-330-4056 or +1-805-330-4056 for Quick Help &…
Attention 설명 없음
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Discriminative Fine-Tuning Discriminative Fine-Tuning is a fine-tuning strategy that is used for ULMFiT type models. Instead of using the same learning rate…
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Cosine Annealing Cosine Annealing is a type of learning rate schedule that has the effect of starting with a large learning rate that is relatively rapidly decreased to a minimum value before…

Similar Papers 제목 키워드 기반

Can Model Merging Improve Aggregation in DiLoCo?

2026-07-03 · Stefan Horoi, Benjamin Thérien, Guy Wolf, Eugene Belilovsky arxiv

Model merging techniques, which aggregate independently finetuned models into one to combine their capabilities, have become a topic of significant interest in recent years, with a broad array of methods having been prop…

Distributed Optimization

Asynchronous Local-SGD Training for Language Modeling

2024-01-17 · Bo Liu, Rachita Chhaparia, Arthur Douillard, Satyen Kale 외

Local stochastic gradient descent (Local-SGD), also referred to as federated averaging, is an approach to distributed optimization where each device performs more than one SGD update per communication. This work presents…

Distributed OptimizationLanguage ModelingLanguage Modelling

Improved Convergence in Parameter-Agnostic Error Feedback through Momentum

2025-11-18 · Abdurakhmon Sadiev, Yury Demidovich, Igor Sokolov, Grigory Malinovsky 외 arxiv

Communication compression is essential for scalable distributed training of modern machine learning models, but it often degrades convergence due to the noise it introduces. Error Feedback (EF) mechanisms are widely adop…

Trade-offs of Local SGD at Scale: An Empirical Study

2021-10-15 · Jose Javier Gonzalez Ortiz, Jonathan Frankle, Mike Rabbat, Ari Morcos 외

As datasets and models become increasingly large, distributed training has become a necessary component to allow deep neural networks to train in reasonable amounts of time. However, distributed training can have substan…

image-classificationImage Classification

Distributed Low-Communication Training with Decoupled Momentum Optimization

2025-10-03 · Sasho Nedelkoski, Alexander Acker, Odej Kao, Soeren Becker 외 arxiv

The training of large models demands substantial computational resources, typically available only in data centers with high-bandwidth interconnects. However, reducing the reliance on high-bandwidth interconnects between…