paper-with-me

홈 › Papers

Memory-Efficient Adaptive Optimization

2019-01-30 · Rohan Anil, Vineet Gupta, Tomer Koren, Yoram Singer

Adaptive gradient-based optimizers such as Adagrad and Adam are crucial for achieving state-of-the-art performance in machine translation and language modeling. However, these methods maintain second-order statistics for each parameter, thus introducing significant memory overheads that restrict the size of the model being used as well as the number of examples in a mini-batch. We describe an effective and flexible adaptive optimization method with greatly reduced memory overhead. Our method retains the benefits of per-parameter adaptivity while allowing significantly larger models and batch sizes. We give convergence guarantees for our method, and demonstrate its effectiveness in training very large translation and language models with up to 2-fold speedups compared to the state-of-the-art.

📄 PDF Abstract BibTeX arXiv:1901.11150

Code (4)

google-research/google-research/tree/master/sm3 공식 구현 jax
emrekuecuek/COMP541
enealor/pytorch-sm3 pytorch
kucukemre96/COMP541

Tasks

Language ModelingLanguage ModellingMachine TranslationTranslation

Methods 이 논문이 사용한 방법론

SM3 설명 없음
AdaGrad AdaGrad is a stochastic optimization method that adapts the learning rate to the parameters. It performs smaller updates for parameters associated with frequently occurring…
Adam 설명 없음

Similar Papers 제목 키워드 기반

AdaLomo: Low-memory Optimization with Adaptive Learning Rate

2023-10-16 · Kai Lv, Hang Yan, Qipeng Guo, Haijun Lv 외

Large language models have achieved remarkable success, but their extensive parameter size necessitates substantial memory for training, thereby setting a high threshold. While the recently proposed low-memory optimizati…

On Adaptivity in Zeroth-Order Optimization

2026-05-05 · Hassan Dbouk, Nidham Gazagnadou, Matthias Reisser, Christos Louizos arxiv

We investigate the effectiveness of adaptive zeroth-order (ZO) optimization for memory-constrained fine-tuning of large language models (LLMs). Contrary to prior claims, we show that adaptive ZO methods such as ZO-Adam o…

Memory Efficient Adaptive Optimization

2019-12-01 · NeurIPS 2019 12 · Rohan Anil, Vineet Gupta, Tomer Koren, Yoram Singer

Adaptive gradient-based optimizers such as Adagrad and Adam are crucial for achieving state-of-the-art performance in machine translation and language modeling. However, these methods maintain second-order statistics for…

Language ModelingLanguage ModellingMachine TranslationTranslation

ReLMXEL: Adaptive RL-Based Memory Controller with Explainable Energy and Latency Optimization

2026-03-18 · Panuganti Chirag Sai, Gandholi Sarat, R. Raghunatha Sarma, Venkata Kalyan Tavva 외 arxiv

Reducing latency and energy consumption is critical to improving the efficiency of memory systems in modern computing. This work introduces ReLMXEL (Reinforcement Learning for Memory Controller with Explainable Energy an…

Reinforcement Learning

Memory-Efficient Optimization with Factorized Hamiltonian Descent

2024-06-14 · Son Nguyen, Lizhang Chen, Bo Liu, Qiang Liu

Modern deep learning heavily depends on adaptive optimizers such as Adam and its variants, which are renowned for their capacity to handle model scaling and streamline hyperparameter tuning. However, these algorithms typ…