Summary of Bones: a Benchmark For Neural Estimation Of Shapley Values, by Davide Napolitano et al.
BONES: a Benchmark fOr Neural Estimation of Shapley values
by Davide Napolitano, Luca Cagliero
First submitted to arxiv on: 23 Jul 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
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 abstract presents a new benchmark, BONES, for evaluating and comparing explainable AI (XAI) models. The focus is on neural estimation of Shapley Values, which quantify feature contributions to model outcomes. Current experiments with neural estimators are difficult to replicate due to a lack of standardization in algorithm implementations, evaluators, and visualizations. To address this gap, BONES provides state-of-the-art neural and traditional estimators, benchmark datasets, modules for training black-box models, and functions for computing evaluation metrics and visualizing results. The authors demonstrate the use of BONES on tabular and image data, making XAI model usage, evaluation, and comparison more straightforward. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper presents a new tool called BONES that helps people understand how artificial intelligence (AI) works. Right now, it’s hard to compare different AI models because they’re not all using the same methods or testing them in the same way. The authors created BONES to solve this problem by providing a collection of popular AI models, common datasets, and tools for training and evaluating these models. They also show how to use BONES on pictures and text data. |
