Summary of A Bilevel Optimization Framework For Imbalanced Data Classification, by Karen Medlin et al.
A Bilevel Optimization Framework for Imbalanced Data Classification
by Karen Medlin, Sven Leyffer, Krishnan Raghavan
First submitted to arxiv on: 15 Oct 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Optimization and Control (math.OC); 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 authors propose a new undersampling approach for addressing imbalanced data issues, which avoids pitfalls of noise and overlap from synthetic data and under-fitting caused by random undersampling. The method rejects majority datapoints that are redundant to accepted datapoints based on their ability to improve model loss, resulting in an optimal subset of training data for classification. This approach is motivated by a bilevel optimization problem formulated to identify the optimal training set. Experimental results show F1 scores up to 10% higher than state-of-the-art methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The researchers developed a new way to handle unbalanced data, which is important because it can affect how well machine learning models work. Their method gets rid of some majority data that’s not helpful for training the model, and keeps the most useful data. This helps the model make better predictions. The results show that this approach is more effective than others in similar situations. |
Keywords
» Artificial intelligence » Classification » Machine learning » Optimization » Synthetic data
