The Explanation Game: Explaining Machine Learning Models with Cooperative Game Theory
@article{Merrick2019TheEG, title={The Explanation Game: Explaining Machine Learning Models with Cooperative Game Theory}, author={Luke Merrick and Ankur Taly}, journal={ArXiv}, year={2019}, volume={abs/1909.08128} }
A number of techniques have been proposed to explain a machine learning (ML) model's prediction by attributing it to the corresponding input features. Popular among these are techniques that apply the Shapley value method from cooperative game theory. While existing papers focus on the axiomatic motivation of Shapley values, and efficient techniques for computing them, they neither justify the game formulations used nor address the uncertainty implicit in their methods' outputs. For instance… CONTINUE READING
Supplemental Code
Github Repo
Via Papers with Code
Supplemental Material for The Explanation Game
Figures, Tables, and Topics from this paper
11 Citations
Causal Shapley Values: Exploiting Causal Knowledge to Explain Individual Predictions of Complex Models
- Computer Science
- NeurIPS
- 2020
- 1
- PDF
An exploration of the influence of path choice in game-theoretic attribution algorithms
- Computer Science, Mathematics
- ArXiv
- 2020
- PDF
Problems with Shapley-value-based explanations as feature importance measures
- Computer Science, Mathematics
- ICML
- 2020
- 18
- Highly Influenced
- PDF
Explaining the data or explaining a model? Shapley values that uncover non-linear dependencies
- Mathematics, Computer Science
- ArXiv
- 2020
- 2
- PDF
Explaining by Removing: A Unified Framework for Model Explanation
- Computer Science, Mathematics
- ArXiv
- 2020
- 1
- Highly Influenced
- PDF
A Projected Stochastic Gradient Algorithm for Estimating Shapley Value Applied in Attribute Importance
- Computer Science
- CD-MAKE
- 2020
Feature Removal Is a Unifying Principle for Model Explanation Methods
- Computer Science, Mathematics
- ArXiv
- 2020
- PDF
Rapid Trust Calibration through Interpretable and Uncertainty-Aware AI
- Computer Science, Medicine
- Patterns
- 2020
- 4
References
SHOWING 1-10 OF 38 REFERENCES
Explaining individual predictions when features are dependent: More accurate approximations to Shapley values
- Mathematics, Computer Science
- ArXiv
- 2019
- 45
- PDF
L-Shapley and C-Shapley: Efficient Model Interpretation for Structured Data
- Computer Science, Mathematics
- ICLR
- 2019
- 54
- PDF
Bounding the Estimation Error of Sampling-based Shapley Value Approximation With/Without Stratifying
- Computer Science, Mathematics
- ArXiv
- 2013
- 50
- PDF
Explaining Deep Neural Networks with a Polynomial Time Algorithm for Shapley Values Approximation
- Computer Science, Mathematics
- ICML
- 2019
- 50
- PDF
An Efficient Explanation of Individual Classifications using Game Theory
- Computer Science, Mathematics
- J. Mach. Learn. Res.
- 2010
- 217
- Highly Influential
- PDF
"Why Should I Trust You?": Explaining the Predictions of Any Classifier
- Computer Science, Mathematics
- HLT-NAACL Demos
- 2016
- 3,853
- PDF
A Unified Approach to Interpreting Model Predictions
- Computer Science, Mathematics
- NIPS
- 2017
- 1,861
- Highly Influential
- PDF
Consistent Individualized Feature Attribution for Tree Ensembles
- Computer Science, Mathematics
- ArXiv
- 2018
- 259
- Highly Influential
- PDF