Summary of Diffusion-augmented Coreset Expansion For Scalable Dataset Distillation, by Ali Abbasi et al.
Diffusion-Augmented Coreset Expansion for Scalable Dataset Distillation
by Ali Abbasi, Shima Imani, Chenyang An, Gayathri Mahalingam, Harsh Shrivastava, Maurice Diesendruck, Hamed Pirsiavash, Pramod Sharma, Soheil Kolouri
First submitted to arxiv on: 5 Dec 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI)
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 two-stage solution for dataset distillation condenses information from extensive datasets into a compact set of synthetic samples by solving a bilevel optimization problem. It first compresses the dataset by selecting only the most informative patches to form a coreset, and then leverages a generative foundation model to dynamically expand this compressed set in real-time. This approach demonstrates significant improvement over state-of-the-art methods on several large-scale dataset distillation benchmarks. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A new way of reducing the amount of data needed to train artificial intelligence models is being developed. Currently, when we try to make AI more efficient, it can take a long time and use lots of energy. This new method makes it faster and uses less energy. It does this by finding the most important parts of the big dataset and then adding some extra information to make it even better. The result is that the AI model learns faster and performs better than before. |
Keywords
» Artificial intelligence » Distillation » Optimization
