Large-scale Reinforcement Learning for Diffusion Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Yinan, Tzeng, Eric, Du, Yilun, Kislyuk, Dmitry
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910305815101440
author Zhang, Yinan
Tzeng, Eric
Du, Yilun
Kislyuk, Dmitry
author_facet Zhang, Yinan
Tzeng, Eric
Du, Yilun
Kislyuk, Dmitry
contents Text-to-image diffusion models are a class of deep generative models that have demonstrated an impressive capacity for high-quality image generation. However, these models are susceptible to implicit biases that arise from web-scale text-image training pairs and may inaccurately model aspects of images we care about. This can result in suboptimal samples, model bias, and images that do not align with human ethics and preferences. In this paper, we present an effective scalable algorithm to improve diffusion models using Reinforcement Learning (RL) across a diverse set of reward functions, such as human preference, compositionality, and fairness over millions of images. We illustrate how our approach substantially outperforms existing methods for aligning diffusion models with human preferences. We further illustrate how this substantially improves pretrained Stable Diffusion (SD) models, generating samples that are preferred by humans 80.3% of the time over those from the base SD model while simultaneously improving both the composition and diversity of generated samples.
format Preprint
id arxiv_https___arxiv_org_abs_2401_12244
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Large-scale Reinforcement Learning for Diffusion Models
Zhang, Yinan
Tzeng, Eric
Du, Yilun
Kislyuk, Dmitry
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Text-to-image diffusion models are a class of deep generative models that have demonstrated an impressive capacity for high-quality image generation. However, these models are susceptible to implicit biases that arise from web-scale text-image training pairs and may inaccurately model aspects of images we care about. This can result in suboptimal samples, model bias, and images that do not align with human ethics and preferences. In this paper, we present an effective scalable algorithm to improve diffusion models using Reinforcement Learning (RL) across a diverse set of reward functions, such as human preference, compositionality, and fairness over millions of images. We illustrate how our approach substantially outperforms existing methods for aligning diffusion models with human preferences. We further illustrate how this substantially improves pretrained Stable Diffusion (SD) models, generating samples that are preferred by humans 80.3% of the time over those from the base SD model while simultaneously improving both the composition and diversity of generated samples.
title Large-scale Reinforcement Learning for Diffusion Models
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2401.12244