Algorithms to estimate Shapley value feature attributions

By Hugh Chen, Ian Covert, Scott M. Lundberg, Su-In Lee. Nature Machine Intelligence, 2022

466 citations recorded by the source corpus.

Feature attributions based on the Shapley value are popular for explaining machine learning models. However, their estimation is complex from both theoretical and computational standpoints. We disentangle this complexity into two main factors: the approach to removing feature information and the tractable estimation strategy. These two factors provide a natural lens through which we can better understand and compare 24 distinct algorithms. Based on the various feature-removal approaches, we describe the multiple types of Shapley value feature attributions and the methods to calculate each one. Then, based on the tractable estimation strategies, we characterize two distinct families of approaches: model-agnostic and model-specific approximations. For the model-agnostic approximations, we benchmark a wide class of estimation approaches and tie them to alternative yet equivalent characterizations of the Shapley value. For the model-specific approximations, we clarify the assumptions crucial to each method’s tractability for linear, tree and deep models. Finally, we identify gaps in the literature and promising future research directions. There are numerous algorithms for generating Shapley value explanations. The authors provide a comprehensive survey of Shapley value feature attribution algorithms by disentangling and clarifying the fundamental challenges underlying their computation.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/c935ad16b132f526b289e16e84d73c5821bfc100

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Fan Yang
2. Ninghao Liu
3. Mengnan Du
4. Dawei Yin
5. Rosanne Liu
6. Ali Farhadi
7. Huiqi Deng
8. Shuaiqiang Wang

Citations in the corpus

• What does a platypus look like? Generating customized prompts for zero-shot image classification (2022, methodology): “In order to measure the importance of regions in the image, we invoke Shapley values [49], a tool from game theory that has become popular for understanding which input information contributes to a model’s final prediction [31, 9].”
• Explainability for Large Language Models: A Survey (2023, background): “This category encompasses four main streams of approaches for generating explanations including feature attribution-based explanation, attention-based explanation, example-based explanation, and natu-ral language explanation (see Figure 3).”