English
Related papers

Related papers: DeMo: Decoupled Momentum Optimization

200 papers

With the increase in the amount of data and the expansion of model scale, distributed parallel training becomes an important and successful technique to address the optimization challenges. Nevertheless, although distributed stochastic…

Machine Learning · Computer Science 2019-09-23 Shuheng Shen , Linli Xu , Jingchang Liu , Xianfeng Liang , Yifei Cheng

Language Model pre-training uses broad data mixtures to enhance performance across domains and languages. However, training on such heterogeneous text corpora requires extensive and expensive efforts. Since these data sources vary…

Machine Learning · Computer Science 2025-04-08 Alex Iacob , Lorenzo Sani , Meghdad Kurmanji , William F. Shen , Xinchi Qiu , Dongqi Cai , Yan Gao , Nicholas D. Lane

We propose an efficient protocol for decentralized training of deep neural networks from distributed data sources. The proposed protocol allows to handle different phases of model training equally well and to quickly adapt to concept…

Machine Learning · Computer Science 2018-11-14 Michael Kamp , Linara Adilova , Joachim Sicking , Fabian Hüger , Peter Schlicht , Tim Wirtz , Stefan Wrobel

As language models scale to trillions of parameters, distributed training across many GPUs becomes essential, yet gradient synchronization over high-bandwidth, low-latency networks remains a critical bottleneck. While recent methods like…

Machine Learning · Computer Science 2025-12-17 Bhavesh Kumar , Roger Jin , Jeffrey Quesnelle

The distributed training of foundation models, particularly large language models (LLMs), demands a high level of communication. Consequently, it is highly dependent on a centralized cluster with fast and reliable interconnects. Can we…

Machine Learning · Computer Science 2025-06-27 Ji Qi , WenPeng Zhu , Li Li , Ming Wu , YingJun Wu , Wu He , Xun Gao , Jason Zeng , Michael Heinrich

Decentralized nonconvex optimization has received increasing attention in recent years in machine learning due to its advantages in system robustness, data privacy, and implementation simplicity. However, three fundamental challenges in…

Machine Learning · Computer Science 2021-05-20 Xin Zhang , Jia Liu , Zhengyuan Zhu , Elizabeth S. Bentley

Optimization algorithms with momentum, e.g., (ADAM), have been widely used for building deep learning models due to the faster convergence rates compared with stochastic gradient descent (SGD). Momentum helps accelerate SGD in the relevant…

Machine Learning · Computer Science 2020-01-24 Jiyang Bai , Yuxiang Ren , Jiawei Zhang

Mixture-of-experts (MoE) architectures enable trillion-parameter LLMs with sparsely activated experts. Expert parallelism (EP) is a widely adopted MoE training strategy, but it suffers from severe all-to-all communication bottlenecks, which…

Channel denoising is a practical and effective technique for mitigating channel estimation errors in multiple-input multiple-output orthogonal frequency-division multiplexing (MIMO-OFDM) systems. However, adapting denoising techniques to…

Signal Processing · Electrical Eng. & Systems 2025-08-14 Sungyoung Ha , Ikbeom Lee , Seunghyeon Jeon , Yo-Seb Jeon

In this paper, we study unconstrained distributed optimization strongly convex problems, in which the exchange of information in the network is captured by a directed graph topology over digital channels that have limited capacity (and…

Systems and Control · Electrical Eng. & Systems 2023-09-12 Apostolos I. Rikos , Wei Jiang , Themistoklis Charalambous , Karl H. Johansson

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…

Machine Learning · Computer Science 2025-03-11 Shuhua Yu , Ding Zhou , Cong Xie , An Xu , Zhi Zhang , Xin Liu , Soummya Kar

Finite-time optimal feedback control for flow networks under information constraints is studied. By utilizing the framework of multi-parametric linear programming, it is demonstrated that when cost/constraints can be modeled or approximated…

Systems and Control · Computer Science 2019-09-24 Saeid Jafari , Ketan Savla

In this paper, we propose DeMuon, a method for decentralized matrix optimization over a given communication topology. DeMuon incorporates matrix orthogonalization via Newton-Schulz iterations-a technique inherited from its centralized…

Optimization and Control · Mathematics 2025-10-03 Chuan He , Shuyi Ren , Jingwei Mao , Erik G. Larsson

Decentralized distributed learning is the key to enabling large-scale machine learning (training) on edge devices utilizing private user-generated local data, without relying on the cloud. However, the practical realization of such…

Machine Learning · Computer Science 2022-09-13 Sai Aparna Aketi , Sangamesh Kodge , Kaushik Roy

Training LLMs relies on distributed implementations using multiple GPUs to compute gradients in parallel with sharded optimizers. However, synchronizing gradients in data parallel setups introduces communication overhead that grows with the…

Machine Learning · Computer Science 2025-10-15 Adel Nabli , Louis Fournier , Pierre Erbacher , Louis Serrano , Eugene Belilovsky , Edouard Oyallon

The fundamental success of large language models hinges upon the efficacious implementation of large-scale distributed training techniques. Nevertheless, building a vast, high-performance cluster featuring high-speed communication…

Computation and Language · Computer Science 2024-01-30 Weigao Sun , Zhen Qin , Weixuan Sun , Shidi Li , Dong Li , Xuyang Shen , Yu Qiao , Yiran Zhong

We consider the problem of decentralized deep learning where multiple agents collaborate to learn from a distributed dataset. While there exist several decentralized deep learning approaches, the majority consider a central parameter-server…

Machine Learning · Computer Science 2020-12-01 Aditya Balu , Zhanhong Jiang , Sin Yong Tan , Chinmay Hedge , Young M Lee , Soumik Sarkar

Regional energy caps limit the growth of any single data center used for large-scale model training. This single-center training paradigm works when model size remains manageable, but exponential growth in the model size and computational…

Machine Learning · Computer Science 2025-12-18 Rongwei Lu , Jingyan Jiang , Chunyang Li , Xingguang Wei , Zhi Wang

We are concerned with decentralized optimization over a compact submanifold, where the loss functions of local datasets are defined by their respective local datasets. A key challenge in decentralized optimization is mitigating the…

Optimization and Control · Mathematics 2024-07-15 Jiang Hu , Kangkang Deng

Large language models (LLM) have become a critical component in many applications of machine learning. However, standard approaches to training LLM require a large number of tightly interconnected accelerators, with devices exchanging…