The many Shapley values for model explanation

By Mukund Sundararajan, Amir-Homayoon Najmi. 2019

847 citations recorded by the source corpus.

The Shapley value has become a popular method to attribute the prediction of a machine-learning model on an input to its base features. The use of the Shapley value is justified by citing [16] showing that it is the \emph{unique} method that satisfies certain good properties (\emph{axioms}). There are, however, a multiplicity of ways in which the Shapley value is operationalized in the attribution problem. These differ in how they reference the model, the training data, and the explanation context. These give very different results, rendering the uniqueness result meaningless. Furthermore, we find that previously proposed approaches can produce counterintuitive attributions in theory and in practice---for instance, they can assign non-zero attributions to features that are not even referenced by the model. In this paper, we use the axiomatic approach to study the differences between some of the many operationalizations of the Shapley value for attribution, and propose a technique called Baseline Shapley (BShap) that is backed by a proper uniqueness result. We also contrast BShap with Integrated Gradients, another extension of Shapley value to the continuous setting.

Subfields: Artificial Intelligence, Safety, Robustness & Interpretability, Game & Decision Theory

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. Su-In Lee
  2. Percy Liang
  3. Mohit Bansal
  4. Scott M. Lundberg
  5. Yoshua Bengio
  6. Luke S. Zettlemoyer
  7. Jianfeng Gao
  8. Noah A. Smith

In Safety, Robustness & Interpretability

  1. Ankur Taly
  2. Christoph Molnar
  3. Bernd Bischl
  4. Su-In Lee
  5. Scott M. Lundberg
  6. Giuseppe Casalicchio
  7. Luke Merrick
  8. Percy Liang

In Game & Decision Theory

  1. Lenon Minorics
  2. Ian Covert
  3. Percy Liang
  4. Bernd Bischl
  5. Dominik Janzing
  6. Tatsunori B. Hashimoto
  7. Su-In Lee
  8. Scott M. Lundberg

Citations in the corpus