Robust ML Auditing using Prior Knowledge

TL;DR

This paper proposes a new method for AI regulation audits that leverages prior knowledge to prevent manipulation by platforms, ensuring more reliable and tamper-proof evaluations.

Contribution

It introduces a formal framework for manipulation-proof auditing using prior knowledge, highlighting limitations of public priors and demonstrating conditions for effective detection.

Findings

Platforms can hide unfairness up to a certain level before detection

Public priors are insufficient for preventing manipulation

Formal conditions enable more robust audits

Abstract

Among the many technical challenges to enforcing AI regulations, one crucial yet underexplored problem is the risk of audit manipulation. This manipulation occurs when a platform deliberately alters its answers to a regulator to pass an audit without modifying its answers to other users. In this paper, we introduce a novel approach to manipulation-proof auditing by taking into account the auditor's prior knowledge of the task solved by the platform. We first demonstrate that regulators must not rely on public priors (e.g. a public dataset), as platforms could easily fool the auditor in such cases. We then formally establish the conditions under which an auditor can prevent audit manipulations using prior knowledge about the ground truth. Finally, our experiments with two standard datasets illustrate the maximum level of unfairness a platform can hide before being detected as malicious. Our formalization and generalization of manipulation-proof auditing with a prior opens up new research directions for more robust fairness audits.