Faculty Recruiting Support CICS

The Shape of Explanations: A Topological Account of Rule-Based Explanations in Machine Learning

30 Nov
Tuesday, 11/30/2021 4:00pm to 5:00pm
Computer Science Building, Room 140
Theory Seminar

Abstract: Rule-based explanations provide simple reasons explaining the behavior of machine learning classifiers at given points in the feature space. Several recent methods (Anchors, LORE, LoRMIkA, etc.) proport to generate rule-based explanations for arbitrary classifiers. But what makes these methods work in general? In this talk, I introduce a topological framework for rule-based explanation methods and provide a characterization of explainability in terms of the definability of a classifier in the explanation topology. I conjecture that classifiers in the wild satisfy this characterization and discuss work in this direction.

The CICS Theory Seminar is free and open to the public. If you are interested in giving a talk, please email Cameron Musco or Rik Sengupta. Note that in addition to being a public lecture series, this is also a one-credit graduate seminar (CompSci 891M) that can be taken repeatedly for credit.