The explanation game: a formal framework for interpretable machine learning

Synthese 198 (10):1–⁠32 (2020)
  Copy   BIBTEX

Abstract

We propose a formal framework for interpretable machine learning. Combining elements from statistical learning, causal interventionism, and decision theory, we design an idealised explanation game in which players collaborate to find the best explanation for a given algorithmic prediction. Through an iterative procedure of questions and answers, the players establish a three-dimensional Pareto frontier that describes the optimal trade-offs between explanatory accuracy, simplicity, and relevance. Multiple rounds are played at different levels of abstraction, allowing the players to explore overlapping causal patterns of variable granularity and scope. We characterise the conditions under which such a game is almost surely guaranteed to converge on a optimal explanation surface in polynomial time, and highlight obstacles that will tend to prevent the players from advancing beyond certain explanatory thresholds. The game serves a descriptive and a normative function, establishing a conceptual space in which to analyse and compare existing proposals, as well as design new and improved solutions.

Analytics

Added to PP
2020-04-03

Downloads
674 (#43,808)

6 months
237 (#13,926)

Historical graph of downloads
How can I increase my downloads?

Author Profiles

Luciano Floridi
Yale University
David Watson
University College London

References found in this work

Thinking, Fast and Slow.Daniel Kahneman - 2011 - New York: New York: Farrar, Straus and Giroux.
Minds, brains, and programs.John Searle - 1980 - Behavioral and Brain Sciences 3 (3):417-57.
Causality: Models, Reasoning and Inference.Judea Pearl - 2000 - New York: Cambridge University Press.
Depth: An Account of Scientific Explanation.Michael Strevens - 2008 - Cambridge: Harvard University Press.

View all 54 references / Add more references