ExpProof : Operationalizing Explanations for Confidential Models with ZKPs

Chhavi Yadav, Evan Laufer, Dan Boneh, Kamalika Chaudhuri

Track: Long Paper Track (up to 9 pages)

Keywords: Explanation, Trust, LIME, Zero-Knowledge Proofs, Auditing

Abstract:

In principle, explanations are intended as a way to increase trust in machine learn-
ing models and are often obligated by regulations. However, many circumstances
where these are demanded are adversarial in nature, meaning the involved parties
have misaligned interests and are incentivized to manipulate explanations for their
purpose. As a result, explainability methods fail to be operational in such settings
despite the demand Bordt et al. (2022). In this paper, we take a step towards op-
erationalizing explanations in adversarial scenarios with Zero-Knowledge Proofs
(ZKPs), a cryptographic primitive. Specifically we explore ZKP-amenable ver-
sions of the popular explainability algorithm LIME and evaluate their performance
on Neural Networks and Random Forests. Our code is publicly available at : https://github.com/infinite-pursuits/ExpProof.

Submission Number: 50