Using Human Feedback to Fine-tune Diffusion Models without Any Reward Model

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yang, Kai, Tao, Jian, Lyu, Jiafei, Ge, Chunjiang, Chen, Jiaxin, Li, Qimai, Shen, Weihan, Zhu, Xiaolong, Li, Xiu
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909147468922880
author Yang, Kai
Tao, Jian
Lyu, Jiafei
Ge, Chunjiang
Chen, Jiaxin
Li, Qimai
Shen, Weihan
Zhu, Xiaolong
Li, Xiu
author_facet Yang, Kai
Tao, Jian
Lyu, Jiafei
Ge, Chunjiang
Chen, Jiaxin
Li, Qimai
Shen, Weihan
Zhu, Xiaolong
Li, Xiu
contents Using reinforcement learning with human feedback (RLHF) has shown significant promise in fine-tuning diffusion models. Previous methods start by training a reward model that aligns with human preferences, then leverage RL techniques to fine-tune the underlying models. However, crafting an efficient reward model demands extensive datasets, optimal architecture, and manual hyperparameter tuning, making the process both time and cost-intensive. The direct preference optimization (DPO) method, effective in fine-tuning large language models, eliminates the necessity for a reward model. However, the extensive GPU memory requirement of the diffusion model's denoising process hinders the direct application of the DPO method. To address this issue, we introduce the Direct Preference for Denoising Diffusion Policy Optimization (D3PO) method to directly fine-tune diffusion models. The theoretical analysis demonstrates that although D3PO omits training a reward model, it effectively functions as the optimal reward model trained using human feedback data to guide the learning process. This approach requires no training of a reward model, proving to be more direct, cost-effective, and minimizing computational overhead. In experiments, our method uses the relative scale of objectives as a proxy for human preference, delivering comparable results to methods using ground-truth rewards. Moreover, D3PO demonstrates the ability to reduce image distortion rates and generate safer images, overcoming challenges lacking robust reward models. Our code is publicly available at https://github.com/yk7333/D3PO.
format Preprint
id arxiv_https___arxiv_org_abs_2311_13231
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Using Human Feedback to Fine-tune Diffusion Models without Any Reward Model
Yang, Kai
Tao, Jian
Lyu, Jiafei
Ge, Chunjiang
Chen, Jiaxin
Li, Qimai
Shen, Weihan
Zhu, Xiaolong
Li, Xiu
Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
Using reinforcement learning with human feedback (RLHF) has shown significant promise in fine-tuning diffusion models. Previous methods start by training a reward model that aligns with human preferences, then leverage RL techniques to fine-tune the underlying models. However, crafting an efficient reward model demands extensive datasets, optimal architecture, and manual hyperparameter tuning, making the process both time and cost-intensive. The direct preference optimization (DPO) method, effective in fine-tuning large language models, eliminates the necessity for a reward model. However, the extensive GPU memory requirement of the diffusion model's denoising process hinders the direct application of the DPO method. To address this issue, we introduce the Direct Preference for Denoising Diffusion Policy Optimization (D3PO) method to directly fine-tune diffusion models. The theoretical analysis demonstrates that although D3PO omits training a reward model, it effectively functions as the optimal reward model trained using human feedback data to guide the learning process. This approach requires no training of a reward model, proving to be more direct, cost-effective, and minimizing computational overhead. In experiments, our method uses the relative scale of objectives as a proxy for human preference, delivering comparable results to methods using ground-truth rewards. Moreover, D3PO demonstrates the ability to reduce image distortion rates and generate safer images, overcoming challenges lacking robust reward models. Our code is publicly available at https://github.com/yk7333/D3PO.
title Using Human Feedback to Fine-tune Diffusion Models without Any Reward Model
topic Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2311.13231