Summary of Bws: Best Window Selection Based on Sample Scores For Data Pruning Across Broad Ranges, by Hoyong Choi et al.
BWS: Best Window Selection Based on Sample Scores for Data Pruning across Broad Ranges
by Hoyong Choi, Nohyun Ki, Hye Won Chung
First submitted to arxiv on: 5 Jun 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Machine Learning (stat.ML)
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 paper introduces Best Window Selection (BWS), a universal and efficient data subset selection method that consistently achieves competitive performance across a broad range of selection ratios. BWS proposes a method to choose the best window subset from samples ordered based on their difficulty scores, allowing for flexibility in choosing window intervals spanning easy to difficult samples. The approach also includes an efficient mechanism for selecting the best window subset by evaluating its quality using kernel ridge regression. Experimental results demonstrate the superior performance of BWS compared to other baselines across a range of selection ratios and datasets including CIFAR-10/100, ImageNet, and training scenarios involving random initialization or fine-tuning of pre-trained models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper helps machines learn better by finding the best part of a big dataset. This is important because big datasets are hard to work with, but smaller parts can be just as good. The method, called Best Window Selection (BWS), picks the best part based on how easy or hard each sample is. It’s flexible and works well in different situations, like when training a new model from scratch or building on an existing one. The results show that BWS does better than other methods in many cases. |
Keywords
» Artificial intelligence » Fine tuning » Regression
