Summary of Gift: Unlocking Full Potential Of Labels in Distilled Dataset at Near-zero Cost, by Xinyi Shang et al.
GIFT: Unlocking Full Potential of Labels in Distilled Dataset at Near-zero Cost
by Xinyi Shang, Peng Sun, Tao Lin
First submitted to arxiv on: 23 May 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Machine Learning (cs.LG)
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 This paper introduces a novel approach to dataset distillation, leveraging soft labels generated by pre-trained teacher models. The authors conduct a comprehensive comparison of various loss functions for soft label utilization and reveal that the choice of loss function has a significant impact on model performance. They propose GIFT, a simple yet effective plug-and-play approach that incorporates soft label refinement and a cosine similarity-based loss function to efficiently utilize full label information. Experimental results show that GIFT consistently enhances state-of-the-art dataset distillation methods across various dataset scales without incurring additional computational costs. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Dataset distillation uses soft labels generated by pre-trained teacher models to train student models. The paper compares different loss functions for using these soft labels and finds that the choice of loss function matters. It proposes a new approach called GIFT, which refines soft labels and uses a special kind of math problem to help the model learn better. This makes the model perform better on other tasks without needing more computer power. |
Keywords
» Artificial intelligence » Cosine similarity » Distillation » Loss function
