Can We Find Strong Lottery Tickets in Generative Models?

TL;DR

This paper demonstrates the existence of strong lottery tickets in generative models, showing that small subnetworks can match or outperform full models without training, using a novel moment-matching score method.

Contribution

It introduces a new method to find strong lottery tickets in generative models, addressing previous challenges of high training costs and performance issues.

Findings

Subnetwork performs similarly or better than dense model with only 10% weights

First demonstration of strong lottery tickets in generative models

Proposed algorithm finds tickets stably and efficiently

Abstract

Yes. In this paper, we investigate strong lottery tickets in generative models, the subnetworks that achieve good generative performance without any weight update. Neural network pruning is considered the main cornerstone of model compression for reducing the costs of computation and memory. Unfortunately, pruning a generative model has not been extensively explored, and all existing pruning algorithms suffer from excessive weight-training costs, performance degradation, limited generalizability, or complicated training. To address these problems, we propose to find a strong lottery ticket via moment-matching scores. Our experimental results show that the discovered subnetwork can perform similarly or better than the trained dense model even when only 10% of the weights remain. To the best of our knowledge, we are the first to show the existence of strong lottery tickets in generative models and provide an algorithm to find it stably. Our code and supplementary materials are publicly available.