Classifier-Free Guidance Diffusion Model

Description

Introduction

There has been an explosion in current trend of generative model ranging from GPT, VAE, GAN, diffusion model, and their variations. This is our project exploring the variation of diffusion model where you do not need a classifier jointly trained.

Goals

The goal of this project is to train a diffusion model with linear noise scheduling and without a classifier guidance. Also, we include a guide on how to play around with our model.

How to run

First, upload all files and folder into a folder on google drive. Then runs on Google Colab. We will have detailed instruction later. For now, here is the link to our model's weight. You can download it and upload to a newly created folder called "models": https://drive.google.com/drive/folders/13iYCsLCtw9T9CVS3tAo3WdnHWxzD9YwP?usp=sharing.

Reference

1. Dhariwal, P., & Nichol, A. (2021). Diffusion models beat gans on image synthesis. Advances in neural information processing systems, 34, 8780-8794.

2. Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., Uszkoreit, J., & Houlsby, N. (2021, June 3). An image is worth 16x16 words: Transformers for image recognition at scale.

3. Gupta, P., Ding, B., Guan, C., & Ding, D. (2024). Generative AI: A systematic review using topic modelling techniques. Data and Information Management, 100066.

4. Ho, J., Jain, A., & Abbeel, P. (2020). Denoising diffusion probabilistic models. Advances in neural information processing systems, 33, 6840-6851.

5. Ho, J., & Salimans, T. (2022). Classifier-free diffusion guidance. arXiv preprint arXiv:2207.12598.

6. Nichol, A. Q., & Dhariwal, P. (2021, July). Improved denoising diffusion probabilistic models. In International conference on machine learning (pp. 8162-8171). PMLR.

7. Ronneberger, O., Fischer, P., & Brox, T. (2015, May 18). U-Net: Convolutional Networks for Biomedical Image Segmentation. arXiv.org.

8. Ruiz, N., Li, Y., Jampani, V., Pritch, Y., Rubinstein, M., & Aberman, K. (2023). Dreambooth: Fine tuning text-to-image diffusion models for subject-driven generation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 22500-22510).

9. Sohl-Dickstein, J., Weiss, E., Maheswaranathan, N., & Ganguli, S. (2015, June). Deep unsupervised learning using nonequilibrium thermodynamics. In International conference on machine learning (pp. 2256-2265). PMLR.

10. Salimans, T., Goodfellow, I., Zaremba, W., Cheung, V., Radford, A. and Chen, X. (2016). Improved techniques for training gans. Advances in neural information processing systems, 29.

11. Heusel, M., Ramsauer, H., Unterthiner, T., Nessler, B., & Hochreiter, S. (2017). Gans trained by a two time-scale update rule converge to a local nash equilibrium. Advances in neural information processing systems, 30.