Summary of Drop: Distributionally Robust Data Pruning, by Artem Vysogorets et al.
DRoP: Distributionally Robust Data Pruning
by Artem Vysogorets, Kartik Ahuja, Julia Kempe
First submitted to arxiv on: 8 Apr 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 In this research paper, the authors investigate the impact of data pruning on classification bias in deep learning models. They demonstrate that existing data pruning algorithms can produce highly biased classifiers and propose a new approach called DRoP to mitigate this issue. The authors use theoretical analysis and empirical experiments to show that their method improves worst-class performance while maintaining average performance on standard computer vision benchmarks. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Data pruning is a technique used to remove redundant or uninformative samples from a dataset, which can help deep learning models train faster and perform better. However, the authors found that existing data pruning algorithms can also make trained models biased towards certain classes. They propose a new approach called DRoP to fix this problem and show it works well on standard computer vision benchmarks. |
Keywords
* Artificial intelligence * Classification * Deep learning * Pruning
