paper-with-me

Papers

Joint MoE Scaling Laws: Mixture of Experts Can Be Memory Efficient

2025-02-07 · Jan Ludziejewski, Maciej Pióro, Jakub Krajewski, Maciej Stefaniak, Michał Krutul, Jan Małaśnicki, Marek Cygan, Piotr Sankowski, Kamil Adamczewski, Piotr Miłoś, Sebastian Jaszczur

Mixture of Experts (MoE) architectures have significantly increased computational efficiency in both research and real-world applications of large-scale machine learning models. However, their scalability and efficiency under memory constraints remain relatively underexplored. In this work, we present joint scaling laws for dense and MoE models, incorporating key factors such as the number of active parameters, dataset size, and the number of experts. Our findings provide a principled framework for selecting the optimal MoE configuration under fixed memory and compute budgets. Surprisingly, we show that MoE models can be more memory-efficient than dense models, contradicting conventional wisdom. To derive and validate the theoretical predictions of our scaling laws, we conduct over 280 experiments with up to 2.7B active parameters and up to 5B total parameters. These results offer actionable insights for designing and deploying MoE models in practical large-scale training scenarios.

📄 PDF Abstract BibTeX arXiv:2502.05172

Code (0)

등록된 구현이 없습니다.

Tasks

Computational EfficiencyMixture-of-Experts

Methods 이 논문이 사용한 방법론

MoE 설명 없음

Similar Papers 제목 키워드 기반

Towards a Comprehensive Scaling Law of Mixture-of-Experts

2025-09-28 · Guoliang Zhao, Yuhan Fu, Shuaipeng Li, Xingwu Sun 외 arxiv

Mixture-of-Experts (MoE) models have become the consensus approach for enabling parameter-efficient scaling and cost-effective deployment in large language models. However, existing scaling laws for dense models are inap…

Holistic Scaling Laws for Optimal Mixture-of-Experts Architecture Optimization

2026-03-23 · Weilin Wan, Jingtao Han, Weizhong Zhang, Cheng Jin arxiv

Scaling laws for Large Language Models govern macroscopic resource allocation, yet translating them into precise Mixture-of-Experts (MoE) architectural configurations remains an open problem due to the combinatorially va…

Scaling Laws for Fine-Grained Mixture of Experts

2024-02-12 · Jakub Krajewski, Jan Ludziejewski, Kamil Adamczewski, Maciej Pióro 외

Mixture of Experts (MoE) models have emerged as a primary solution for reducing the computational cost of Large Language Models. In this work, we analyze their scaling properties, incorporating an expanded range of varia…

Mixture-of-Experts

Mixtures of Experts Unlock Parameter Scaling for Deep RL

2024-02-13 · Johan Obando-Ceron, Ghada Sokar, Timon Willi, Clare Lyle 외

The recent rapid progress in (self) supervised learning models is in large part predicted by empirical scaling laws: a model's performance scales proportionally to its size. Analogous scaling laws remain elusive for rein…

reinforcement-learningReinforcement LearningSelf-Supervised Learning

Ultra-Sparse Memory Network

2024-11-19 · Zihao Huang, Qiyang Min, Hongzhi Huang, Defa Zhu 외

It is widely acknowledged that the performance of Transformer models is exponentially related to their number of parameters and computational complexity. While approaches like Mixture of Experts (MoE) decouple parameter …

Mixture-of-Experts