Summary of Shap-select: Lightweight Feature Selection Using Shap Values and Regression, by Egor Kraev et al.
Shap-Select: Lightweight Feature Selection Using SHAP Values and Regression
by Egor Kraev, Baran Koseoglu, Luca Traverso, Mohammed Topiwalla
First submitted to arxiv on: 9 Oct 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
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 proposes a novel feature selection framework called shap-select, designed to improve machine learning model performance by reducing dimensionality in high-dimensional datasets. The framework uses Shapley values to perform linear or logistic regression on the target variable and select features based on the signs and significance levels of the coefficients. Evaluations on the Kaggle credit card fraud dataset demonstrate the effectiveness of shap-select compared to established methods like RFE, HISEL, Boruta, and a simpler Shapley value-based approach. The proposed framework offers interpretability, computational efficiency, and performance, making it a robust solution for tabular regression and classification tasks. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Imagine you have a huge set of data points with lots of features that are all connected to each other. To make sense of this data, we need to choose the most important features that help us understand what’s going on. This paper introduces a new way to do just that called shap-select. It works by looking at how each feature is related to the thing we’re trying to predict and then selecting the ones that are most useful. The authors tested this method on some real-world data and showed it works better than other popular methods. |
Keywords
» Artificial intelligence » Classification » Feature selection » Logistic regression » Machine learning » Regression
