Summary of An Efficient Framework For Crediting Data Contributors Of Diffusion Models, by Chris Lin et al.
An Efficient Framework for Crediting Data Contributors of Diffusion Models
by Chris Lin, Mingyu Lu, Chanwoo Kim, Su-In Lee
First submitted to arxiv on: 9 Jun 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Computer Vision and Pattern Recognition (cs.CV)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed paper addresses the problem of attributing global properties of diffusion models to data contributors. This is crucial in real-world settings where model performance is driven by training data and incentives are needed to encourage sharing quality data. The authors introduce a method to efficiently estimate Shapley values for diffusion models, leveraging model pruning and fine-tuning. The method is evaluated on three use cases: image quality, demographic diversity, and aesthetic quality. Results show that the framework can identify important data contributors across models’ global properties, outperforming existing attribution methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper explores how to give credit to people who share data used to train artificial intelligence (AI) models. When AI models are trained on big datasets, it’s important to know which pieces of data were most important for the model’s performance. The authors develop a new way to do this, called the Shapley value, and show that their method is more accurate than existing methods. |
Keywords
» Artificial intelligence » Diffusion » Fine tuning » Pruning
