Aligning Text-to-Image Diffusion Models with Reward Backpropagation

This is the official implementation of our paper Aligning Text-to-Image Diffusion Models with Reward Backpropagation by Mihir Prabhudesai, Anirudh Goyal, Deepak Pathak, and Katerina Fragkiadaki.

Abstract

Text-to-image diffusion models have recently emerged at the forefront of image generation, powered by very large-scale unsupervised or weakly supervised text-to-image training datasets. Due to the weakly supervised training, controlling their behavior in downstream tasks, such as maximizing human-perceived image quality, image-text alignment, or ethical image generation, is difficult. Recent works finetune diffusion models to downstream reward functions using vanilla reinforcement learning, notorious for the high variance of the gradient estimators. In this paper, we propose AlignProp, a method that aligns diffusion models to downstream reward functions using end-to-end backpropagation of the reward gradient through the denoising process. While naive implementation of such backpropagation would require prohibitive memory resources for storing the partial derivatives of modern text-to-image models, AlignProp finetunes low-rank adapter weight modules and uses gradient checkpointing, to render its memory usage viable. We test AlignProp in finetuning diffusion models to various objectives, such as image-text semantic alignment, aesthetics, compressibility and controllability of the number of objects present, as well as their combinations. We show AlignProp achieves higher rewards in fewer training steps than alternatives, while being conceptually simpler, making it a straightforward choice for optimizing diffusion models for differentiable reward functions of interest.

Code

Installation

Create a conda environment with the following command:

conda create -n alignprop python=3.10
conda activate alignprop
pip install -r requirements.txt

Training Code

Accelerate will automatically handle multi-GPU setting. The code can work on a single GPU, as we automatically handle gradient accumulation as per the available GPUs in the CUDA_VISIBLE_DEVICES environment variable. For our experiments, we used 4 A100s- 40GB RAM to run our code. If you are using a GPU with a smaller RAM, please edit the train_batch_size variable accordingly. Further if u are bottlenecked by GPU memory, consider using AlignProp with K=1, this will significantly reduce the memroy usage.

Aesthetic Reward model.

Currently we early stop the code to prevent overfitting, however feel free to play with the num_epochs variable as per your needs.

bash aesthetic.sh

If you are memory bottlenecked use AlignProp K=1. Lower values of backprop_timestep (i.e higher values of K) can help with focusing on more semantic details:

bash aesthetic_k1.sh

HPSv2 Reward model.

bash hps.sh

Similarly, for low GPU VRAM, K=1 can be used.

bash hps_k1.sh

Acknowledgement

Our codebase is directly built on top of DDPO. We would like to thank Kevin Black and team, for opensourcing their code.

Citation

If you find this work useful in your research, please cite:

@misc{prabhudesai2023aligning,
      title={Aligning Text-to-Image Diffusion Models with Reward Backpropagation}, 
      author={Mihir Prabhudesai and Anirudh Goyal and Deepak Pathak and Katerina Fragkiadaki},
      year={2023},
      eprint={2310.03739},
      archivePrefix={arXiv},
      primaryClass={cs.CV}
}