FastSHAP: Real-Time Shapley Value Estimation
Neil Jethani, Mukund Sudarshan, Ian Connick Covert, Su-In Lee, Rajesh Ranganath
Abstract
Although Shapley values are theoretically appealing for explaining black-box models, they are costly to calculate and thus impractical in settings that involve large, high-dimensional models. To remedy this issue, we introduce FastSHAP, a new method for estimating Shapley values in a single forward pass using a learned explainer model. To enable efficient training without requiring ground truth Shapley values, we develop an approach to train FastSHAP via stochastic gradient descent using a weighted least-squares objective function. In our experiments with tabular and image datasets, we compare FastSHAP to existing estimation approaches and find that it generates accurate explanations with an orders-of-magnitude speedup.
BibTeX
@inproceedings{
jethani2022fastshap,
title={Fast{SHAP}: Real-Time Shapley Value Estimation},
author={Neil Jethani and Mukund Sudarshan and Ian Connick Covert and Su-In Lee and Rajesh Ranganath},
booktitle={International Conference on Learning Representations},
year={2022},
url={https://openreview.net/forum?id=Zq2G_VTV53T}
}