paper-with-me

Papers

Convergence Rates for Softmax Gating Mixture of Experts

2025-03-05 · Huy Nguyen, Nhat Ho, Alessandro Rinaldo

Mixture of experts (MoE) has recently emerged as an effective framework to advance the efficiency and scalability of machine learning models by softly dividing complex tasks among multiple specialized sub-models termed experts. Central to the success of MoE is an adaptive softmax gating mechanism which takes responsibility for determining the relevance of each expert to a given input and then dynamically assigning experts their respective weights. Despite its widespread use in practice, a comprehensive study on the effects of the softmax gating on the MoE has been lacking in the literature. To bridge this gap in this paper, we perform a convergence analysis of parameter estimation and expert estimation under the MoE equipped with the standard softmax gating or its variants, including a dense-to-sparse gating and a hierarchical softmax gating, respectively. Furthermore, our theories also provide useful insights into the design of sample-efficient expert structures. In particular, we demonstrate that it requires polynomially many data points to estimate experts satisfying our proposed \emph{strong identifiability} condition, namely a commonly used two-layer feed-forward network. In stark contrast, estimating linear experts, which violate the strong identifiability condition, necessitates exponentially many data points as a result of intrinsic parameter interactions expressed in the language of partial differential equations. All the theoretical results are substantiated with a rigorous guarantee.

📄 PDF Abstract BibTeX arXiv:2503.03213

Code (0)

등록된 구현이 없습니다.

Tasks

Mixture-of-Expertsparameter estimation

Methods 이 논문이 사용한 방법론

MoE 설명 없음
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Hierarchical Softmax 설명 없음
Adaptive Softmax Adaptive Softmax is a speedup technique for the computation of probability distributions over words. The adaptive softmax is…

Similar Papers 제목 키워드 기반

Sigmoid Gating is More Sample Efficient than Softmax Gating in Mixture of Experts

2024-05-22 · Huy Nguyen, Nhat Ho, Alessandro Rinaldo

The softmax gating function is arguably the most popular choice in mixture of experts modeling. Despite its widespread use in practice, the softmax gating may lead to unnecessary competition among experts, potentially ca…

Mixture-of-Experts

Demystifying Softmax Gating Function in Gaussian Mixture of Experts

2023-05-05 · NeurIPS 2023 11

Understanding the parameter estimation of softmax gating Gaussian mixture of experts has remained a long-standing open problem in the literature. It is mainly due to three fundamental theoretical challenges associated wi…

Mixture-of-Expertsparameter estimation

Statistical Perspective of Top-K Sparse Softmax Gating Mixture of Experts

2023-09-25 · Huy Nguyen, Pedram Akbarian, Fanqi Yan, Nhat Ho

Top-K sparse softmax gating mixture of experts has been widely used for scaling up massive deep-learning architectures without increasing the computational cost. Despite its popularity in real-world applications, the the…

Density EstimationMixture-of-Expertsparameter estimation

A General Theory for Softmax Gating Multinomial Logistic Mixture of Experts

2023-10-22 · Huy Nguyen, Pedram Akbarian, TrungTin Nguyen, Nhat Ho

Mixture-of-experts (MoE) model incorporates the power of multiple submodels via gating functions to achieve greater performance in numerous regression and classification applications. From a theoretical perspective, whil…

Density EstimationMixture-of-Expertsparameter estimationregression

On Bayesian Softmax-Gated Mixture-of-Experts Models

2026-04-22 · Nicola Bariletto, Huy Nguyen, Nhat Ho, Alessandro Rinaldo arxiv

Mixture-of-experts models provide a flexible framework for learning complex probabilistic input-output relationships by combining multiple expert models through an input-dependent gating mechanism. These models have beco…

Density Estimation