0

GAN Cocktail: mixing GANs without dataset access

Today's generative models are capable of synthesizing high-fidelity images, but each model specializes on a specific target domain.

Year
2021
Venue
arXiv 2021
Authors
3
Hosting
Abstract onlyARXIV-DEFAULT

Cite

Notes

Only stored in your browser.

Attribution

Abstract & full text
arxiv.org/abs/2106.03847v2ARXIV-DEFAULT
TL;DR
Semantic Scholar
Attribution policy →

Abstract

Today's generative models are capable of synthesizing high-fidelity images, but each model specializes on a specific target domain. This raises the need for model merging: combining two or more pretrained generative models into a single unified one. In this work we tackle the problem of model merging, given two constraints that often come up in the real world: (1) no access to the original training data, and (2) without increasing the size of the neural network. To the best of our knowledge, model merging under these constraints has not been studied thus far. We propose a novel, two-stage solution. In the first stage, we transform the weights of all the models to the same parameter space by a technique we term model rooting. In the second stage, we merge the rooted models by averaging their weights and fine-tuning them for each specific domain, using only data generated by the original trained models. We demonstrate that our approach is superior to baseline methods and to existing transfer learning techniques, and investigate several applications.

Authors

3