Generating Counterfactual and Contrastive Explanations using SHAP
By Shubham Rathi. 2019
65 citations recorded by the source corpus.
With the advent of GDPR, the domain of explainable AI and model interpretability has gained added impetus. Methods to extract and communicate visibility into decision-making models have become legal requirement. Two specific types of explanations, contrastive and counterfactual have been identified as suitable for human understanding. In this paper, we propose a model agnostic method and its systemic implementation to generate these explanations using shapely additive explanations (SHAP). We discuss a generative pipeline to create contrastive explanations and use it to further to generate counterfactual datapoints. This pipeline is tested and discussed on the IRIS, Wine Quality & Mobile Features dataset. Analysis of the results obtained follows.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Information Systems and Management
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
- Andreas Butz
- John P. Dickerson
- Sahil Verma
- Felicitas Buchner
- Adrian Krüger
- Malin Eiband
- Keegan E. Hines
- Michael Chromik
In Computer Vision and Pattern Recognition
- Andreas Butz
- John P. Dickerson
- Sahil Verma
- Felicitas Buchner
- Adrian Krüger
- Malin Eiband
- Keegan E. Hines
- Michael Chromik
In Information Systems and Management
- Andreas Butz
- John P. Dickerson
- Sahil Verma
- Felicitas Buchner
- Adrian Krüger
- Malin Eiband
- Keegan E. Hines
- Michael Chromik
Citations in the corpus
- A survey of algorithmic recourse: definitions, formulations, solutions, and prospects (2020, neutral): “using SHAP [152] E heuristic query (2019.”
- I Think I Get Your Point, AI! The Illusion of Explanatory Depth in Explainable AI (2021, methodology): “The properties of SHAP enable novel ways for interactivity [13, 43] to provide selective, contrastive, and interactive explanations [32] that might bridge the gap between local and global understanding [42].”
- Counterfactual Explanations and Algorithmic Recourses for Machine Learning: A Review (2020, unknown): “Finally, some approaches do not cast the goal into an optimization problem and solve it using heuristics [126, 173, 254, 334].”
- Counterfactual Explanations for Machine Learning: A Review (2020, methodology): “[91] Black-box Agnostic No No No No No No -”