English
Related papers

Related papers: LoCo: Low-Bit Communication Adaptor for Large-scal…

200 papers

Pre-training large language models (LLMs) increasingly requires distributed compute, yet bandwidth constraints make it difficult to scale beyond well-provisioned datacenters-especially when model parallelism forces frequent, large…

Machine Learning · Computer Science 2026-01-06 Yazan Obeidi , Amir Sarfi , Joel Lidin , Paul Janson , Eugene Belilovsky

Large-scale machine learning training, in particular distributed stochastic gradient descent, needs to be robust to inherent system variability such as node straggling and random communication delays. This work considers a distributed…

Machine Learning · Computer Science 2019-03-08 Jianyu Wang , Gauri Joshi

Due to the substantial computational cost, training state-of-the-art deep neural networks for large-scale datasets often requires distributed training using multiple computation workers. However, by nature, workers need to frequently…

Machine Learning · Computer Science 2018-02-21 Yusuke Tsuzuku , Hiroto Imachi , Takuya Akiba

Communicating information, like gradient vectors, between computing nodes in distributed and federated learning is typically an unavoidable burden, resulting in scalability issues. Indeed, communication might be slow and costly. Recent…

Machine Learning · Computer Science 2020-10-08 Alyazeed Albasyoni , Mher Safaryan , Laurent Condat , Peter Richtárik

The growth of large language models (LLMs) increases challenges of accelerating distributed training across multiple GPUs in different data centers. Moreover, concerns about data privacy and data exhaustion have heightened interest in…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-02-18 Zhenheng Tang , Zichen Tang , Junlin Huang , Xinglin Pan , Rudan Yan , Yuxin Wang , Amelie Chi Zhou , Shaohuai Shi , Xiaowen Chu , Bo Li

We propose Adaptive Compressed Gradient Descent (AdaCGD) - a novel optimization algorithm for communication-efficient training of supervised machine learning models with adaptive compression level. Our approach is inspired by the recently…

Machine Learning · Computer Science 2022-11-02 Maksim Makarenko , Elnur Gasanov , Rustem Islamov , Abdurakhmon Sadiev , Peter Richtarik

Scaling models has led to significant advancements in deep learning, but training these models in decentralized settings remains challenging due to communication bottlenecks. While existing compression techniques are effective in…

Machine Learning · Computer Science 2025-06-03 Sameera Ramasinghe , Thalaiyasingam Ajanthan , Gil Avraham , Yan Zuo , Alexander Long

Modern deep neural networks often require distributed training with many workers due to their large size. As the number of workers increases, communication overheads become the main bottleneck in data-parallel minibatch stochastic gradient…

Machine Learning · Statistics 2024-11-07 Tim Tsz-Kit Lau , Weijian Li , Chenwei Xu , Han Liu , Mladen Kolar

In recent years, large-scale vision-language models (VLMs) have demonstrated remarkable performance on multimodal understanding and reasoning tasks. However, handling high-dimensional visual features often incurs substantial computational…

Computer Vision and Pattern Recognition · Computer Science 2025-12-23 Xiaoyang Guo , Keze Wang

With the increasing scale of machine learning tasks, it has become essential to reduce the communication between computing nodes. Early work on gradient compression focused on the bottleneck between CPUs and GPUs, but…

Optimization and Control · Mathematics 2020-06-18 Sarit Khirirat , Sindri Magnússon , Arda Aytekin , Mikael Johansson

Low-rank adaptation (LoRA) has become the default approach to fine-tune large language models (LLMs) due to its significant reduction in trainable parameters. However, trainable parameter demand for LoRA increases with increasing model…

Computation and Language · Computer Science 2024-06-19 Seyedarmin Azizi , Souvik Kundu , Massoud Pedram

Distributed trajectory optimization via ADMM-DDP is a powerful approach for coordinating multi-agent systems, but it requires extensive tuning of tightly coupled hyperparameters that jointly govern local task performance and global…

Machine Learning · Computer Science 2025-09-08 Bingheng Wang , Yichao Gao , Tianchen Sun , Lin Zhao

Quantized training of Large Language Models (LLMs) remains an open challenge, as maintaining accuracy while performing all matrix multiplications in low precision has proven difficult. This is particularly the case when fine-tuning…

Machine Learning · Computer Science 2025-11-06 Saleh Ashkboos , Mahdi Nikdan , Soroush Tabesh , Roberto L. Castro , Torsten Hoefler , Dan Alistarh

The rapid scaling of large language models (LLMs) has made low-precision training essential for reducing memory, improving efficiency, and enabling larger models and datasets. Existing convergence theories for adaptive optimizers, however,…

Machine Learning · Computer Science 2026-03-03 Xuan Tang , Jichu Li , Difan Zou

Parallel implementations of stochastic gradient descent (SGD) have received significant research attention, thanks to excellent scalability properties of this algorithm, and to its efficiency in the context of training deep neural networks.…

Machine Learning · Computer Science 2017-12-07 Dan Alistarh , Demjan Grubic , Jerry Li , Ryota Tomioka , Milan Vojnovic

It is widely believed that stochastic gradient descent (SGD) performs significantly worse than adaptive optimizers such as Adam in pre-training Large Language Models (LLMs). Yet the underlying reason for this gap remains unclear. In this…

Machine Learning · Computer Science 2026-05-19 Athanasios Glentis , Dawei Li , Chung-Yiu Yau , Mingyi Hong

We propose and study a new class of gradient communication mechanisms for communication-efficient training -- three point compressors (3PC) -- as well as efficient distributed nonconvex optimization algorithms that can take advantage of…

Machine Learning · Computer Science 2022-02-03 Peter Richtárik , Igor Sokolov , Ilyas Fatkhullin , Elnur Gasanov , Zhize Li , Eduard Gorbunov

This paper addresses distributed stochastic optimization problems under non-i.i.d. data, focusing on the inherent trade-offs between communication and computational efficiency. To this end, we propose FlexGT, a flexible snapshot gradient…

Optimization and Control · Mathematics 2026-02-17 Yan Huang , Jinming Xu , Li Chai , Jiming Chen , Karl H. Johansson

Fine-tuning language models (LMs) with the Adam optimizer often demands excessive memory, limiting accessibility. The "in-place" version of Stochastic Gradient Descent (IP-SGD) and Memory-Efficient Zeroth-order Optimizer (MeZO) have been…

Machine Learning · Computer Science 2024-10-10 Zeman Li , Xinwei Zhang , Peilin Zhong , Yuan Deng , Meisam Razaviyayn , Vahab Mirrokni

Training and fine-tuning large language models (LLMs) come with challenges related to memory and computational requirements due to the increasing size of the model weights and the optimizer states. Various techniques have been developed to…

Machine Learning · Computer Science 2025-12-09 Yehonathan Refael , Jonathan Svirsky , Boris Shustin , Wasim Huleihel , Ofir Lindenbaum