Summary of The Distributional Uncertainty Of the Shap Score in Explainable Machine Learning, by Santiago Cifuentes and Leopoldo Bertossi and Nina Pardal and Sergio Abriola and Maria Vanina Martinez and Miguel Romero
The Distributional Uncertainty of the SHAP score in Explainable Machine Learning
by Santiago Cifuentes, Leopoldo Bertossi, Nina Pardal, Sergio Abriola, Maria Vanina Martinez, Miguel Romero
First submitted to arxiv on: 23 Jan 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: Machine Learning (cs.LG); Logic in Computer Science (cs.LO)
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 proposed framework offers a principled approach for reasoning about SHAP scores in the presence of unknown entity population distributions. By defining an uncertainty region containing potential distributions and considering the SHAP score as a function over this region, the framework provides tight ranges for feature scores, enabling more robust scoring. The study’s findings also highlight the NP-completeness of related problems, underscoring the significance of the proposed approach. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper proposes a new way to understand how important each part is in a machine learning model. Right now, we use something called SHAP scores, which can be tricky because they rely on knowing the whole population of things being scored. But what if that’s unknown? The authors suggest a framework for dealing with this uncertainty, which allows us to get a range of possible scores for each part instead of just one number. This could help make feature scoring more reliable and useful. |
Keywords
* Artificial intelligence * Machine learning
