Patch-level Routing in Mixture-of-Experts is Provably Sample-efficient for Convolutional Neural Networks

In deep learning, mixture-of-experts (MoE) activates one or few experts (sub-networks) on a per-sample or per-token basis, resulting in significant computation reduction. The recently proposed patch-level routing in MoE (pMoE) divides each input into n patches (or tokens) and sends l patches (l≪ n) to each expert through prioritized routing. pMoE has demonstrated great empirical success in reducing training and inference costs while maintaining test accuracy. However, the theoretical explanation of pMoE and the general MoE remains elusive. Focusing on a supervised classification task using a mixture of two-layer convolutional neural networks (CNNs), we show for the first time that pMoE provably reduces the required number of training samples to achieve desirable generalization (referred to as the sample complexity) by a factor in the polynomial order of n/l, and outperforms its single-expert counterpart of the same or even larger capacity. The advantage results from the discriminative routing property, which is justified in both theory and practice that pMoE routers can filter label-irrelevant patches and route similar class-discriminative patches to the same expert. Our experimental results on MNIST, CIFAR-10, and CelebA support our theoretical findings on pMoE's generalization and show that pMoE can avoid learning spurious correlations.


page 1

page 2

page 3

page 4


StableMoE: Stable Routing Strategy for Mixture of Experts

The Mixture-of-Experts (MoE) technique can scale up the model size of Tr...

Towards Understanding Mixture of Experts in Deep Learning

The Mixture-of-Experts (MoE) layer, a sparsely-activated model controlle...

On the Representation Collapse of Sparse Mixture of Experts

Sparse mixture of experts provides larger model capacity while requiring...

A Theoretical Understanding of shallow Vision Transformers: Learning, Generalization, and Sample Complexity

Vision Transformers (ViTs) with self-attention modules have recently ach...

SMILE: Scaling Mixture-of-Experts with Efficient Bi-level Routing

The mixture of Expert (MoE) parallelism is a recent advancement that sca...

Soft Merging of Experts with Adaptive Routing

Sparsely activated neural networks with conditional computation learn to...

Deep Mixture of Diverse Experts for Large-Scale Visual Recognition

In this paper, a deep mixture of diverse experts algorithm is developed ...

Please sign up or login with your details

Forgot password? Click here to reset