Patch-level Routing in Mixture-of-Experts is Provably Sample-efficient for Convolutional Neural Networks
Jul 25, 2023
Sprecher:innen
Mohammed Nowaz Rabbani Chowdhury
Řečník · 0 sledujících
Shuai Zhang
Řečník · 1 sledující
Pin-Yu Chen
Řečník · 3 sledující
Über
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 ($l\ll n$) 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.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 ($l\ll n$) 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 exp…
Gefällt euch das Format? Vertraut auf SlidesLive, um euer nächstes Event festzuhalten!
Professionelle Aufzeichnung und Livestreaming – weltweit.
Freigeben
Empfohlene Videos
Präsentationen, deren Thema, Kategorie oder Sprecher:in ähnlich sind
For Pre-Trained Vision Models in Motor Control, Not All Policy Learning Methods are Created Equal
Zhlédnout později
Yingdong Hu, …
Unsupervised Skill Discovery for Learning Shared Structures across Changing Environments
Zhlédnout později
Unscented Autoencoder
Zhlédnout později
Faris Janjoš, …
Loss Balancing for Fair Supervised Learning
Zhlédnout později
Hardness of Independent Learning and Sparse Equilibrium Computation in Markov Games
Zhlédnout později
