Parameters vs FLOPs: Scaling Laws for Optimal Sparsity for Mixture-of-Experts Language Models

• URL: http://arxiv.org/abs/2501.12370v2
• Date: Sat, 25 Jan 2025 02:41:24 GMT
• Title: Parameters vs FLOPs: Scaling Laws for Optimal Sparsity for Mixture-of-Experts Language Models
• Authors: Samira Abnar, Harshay Shah, Dan Busbridge, Alaaeldin Mohamed Elnouby Ali, Josh Susskind, Vimal Thilak,
• Abstract summary: sparse Mixture-of-Experts (MoEs) allow scaling the number of parameters without proportionally increasing the FLOPs per example. We investigate how varying the sparsity level, i.e., the fraction of inactive parameters, impacts model's performance during pretraining and downstream few-shot evaluation.
• Score: 10.517704202614091
• License:
• Abstract: Scaling the capacity of language models has consistently proven to be a reliable approach for improving performance and unlocking new capabilities. Capacity can be primarily defined by two dimensions: the number of model parameters and the compute per example. While scaling typically involves increasing both, the precise interplay between these factors and their combined contribution to overall capacity remains not fully understood. We explore this relationship in the context of sparse Mixture-of-Experts (MoEs), which allow scaling the number of parameters without proportionally increasing the FLOPs per example. We investigate how varying the sparsity level, i.e., the fraction of inactive parameters, impacts model's performance during pretraining and downstream few-shot evaluation. We find that under different constraints (e.g., parameter size and total training compute), there is an optimal level of sparsity that improves both training efficiency and model performance. These results provide a better understanding of the impact of sparsity in scaling laws for MoEs and complement existing works in this area, offering insights for designing more efficient architectures.

Related papers

• Efficient Source-Free Time-Series Adaptation via Parameter Subspace Disentanglement [0.7558576228782637]
  We propose a framework for efficient Source-Free Domain Adaptation (SFDA) Our approach introduces an improved paradigm for source-model preparation and target-side adaptation. We demonstrate that our framework is compatible with various SFDA methods and achieves significant computational efficiency.
  arXiv  Detail & Related papers  (2024-10-03T02:12:03Z)
• Observational Scaling Laws and the Predictability of Language Model Performance [51.2336010244645]
  We propose an observational approach that bypasses model training and instead builds scaling laws from 100 publically available models. We show that several emergent phenomena follow a smooth, sigmoidal behavior and are predictable from small models. We show how to predict the impact of post-training interventions like Chain-of-Thought and Self-Consistency as language model capabilities continue to improve.
  arXiv  Detail & Related papers  (2024-05-17T17:49:44Z)
• Mixtures of Experts Unlock Parameter Scaling for Deep RL [54.26191237981469]
  In this paper, we demonstrate that incorporating Mixture-of-Expert (MoE) modules into value-based networks results in more parameter-scalable models. This work thus provides strong empirical evidence towards developing scaling laws for reinforcement learning.
  arXiv  Detail & Related papers  (2024-02-13T17:18:56Z)
• Parameter-Efficient Sparsity Crafting from Dense to Mixture-of-Experts for Instruction Tuning on General Tasks [5.536630285985836]
  We introduce parameter-efficient sparsity crafting (PESC) PESC crafts dense models into sparse models using the mixture-of-experts (MoE) architecture. Our best sparse model outperforms other sparse and dense models and exhibits superior general capabilities compared to GP3.5.
  arXiv  Detail & Related papers  (2024-01-05T09:58:09Z)
• Can pruning make Large Language Models more efficient? [0.0]
  This paper investigates the application of weight pruning as an optimization strategy for Transformer architectures. Our findings suggest that significant reductions in model size are attainable without considerable compromise on performance. This work seeks to bridge the gap between model efficiency and performance, paving the way for more scalable and environmentally responsible deep learning applications.
  arXiv  Detail & Related papers  (2023-10-06T20:28:32Z)
• Scaling Laws for Sparsely-Connected Foundation Models [70.41266138010657]
  We explore the impact of parameter sparsity on the scaling behavior of Transformers trained on massive datasets. We identify the first scaling law describing the relationship between weight sparsity, number of non-zero parameters, and amount of training data.
  arXiv  Detail & Related papers  (2023-09-15T16:29:27Z)
• Understanding Parameter Sharing in Transformers [53.75988363281843]
  Previous work on Transformers has focused on sharing parameters in different layers, which can improve the performance of models with limited parameters by increasing model depth. We show that the success of this approach can be largely attributed to better convergence, with only a small part due to the increased model complexity. Experiments on 8 machine translation tasks show that our model achieves competitive performance with only half the model complexity of parameter sharing models.
  arXiv  Detail & Related papers  (2023-06-15T10:48:59Z)
• A Generic Performance Model for Deep Learning in a Distributed Environment [0.7829352305480285]
  We propose a generic performance model of an application in a distributed environment with a generic expression of the application execution time. We have evaluated the proposed model on three deep learning frameworks (i.e., MXnet, and Pytorch)
  arXiv  Detail & Related papers  (2023-05-19T13:30:34Z)
• Scaling Pre-trained Language Models to Deeper via Parameter-efficient Architecture [68.13678918660872]
  We design a more capable parameter-sharing architecture based on matrix product operator (MPO) MPO decomposition can reorganize and factorize the information of a parameter matrix into two parts. Our architecture shares the central tensor across all layers for reducing the model size.
  arXiv  Detail & Related papers  (2023-03-27T02:34:09Z)
• MoEfication: Conditional Computation of Transformer Models for Efficient Inference [66.56994436947441]
  Transformer-based pre-trained language models can achieve superior performance on most NLP tasks due to large parameter capacity, but also lead to huge computation cost. We explore to accelerate large-model inference by conditional computation based on the sparse activation phenomenon. We propose to transform a large model into its mixture-of-experts (MoE) version with equal model size, namely MoEfication.
  arXiv  Detail & Related papers  (2021-10-05T02:14:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.