Search papers, labs, and topics across Lattice.
This paper introduces a novel audit protocol that enhances the integrity of algorithmic audits by allowing auditors to query machine learning models in an oblivious manner, thereby mitigating the risk of manipulation by model providers. By employing a Private Information Retrieval mechanism, the protocol ensures that providers cannot anticipate which instances will be audited, making it significantly harder for them to conceal unfairness. Experimental results validate the protocol's effectiveness, demonstrating that it increases the difficulty of manipulation and the likelihood of detection in fairness evaluations.
Auditors can now detect manipulative practices by model providers with a novel oblivious audit protocol that thwarts strategic responses to fairness evaluations.
Audits have emerged as a critical instrument for algorithmic governance, providing a mechanism for external scrutiny and governance of machine learning models. However, ensuring the integrity of such assessments remains a challenging issue. For instance in regulatory contexts, audits are typically declared or easily detected, thus enabling model providers to manipulate the process, whether intentionally or inadvertently. This vulnerability is particularly acute in the context of fairness evaluations, in which providers can often infer sensitive attributes and strategically equalize allocation rates between groups to satisfy fairness metrics. In this paper, we introduce a novel audit protocol designed to significantly increase the post-audit detectability of such manipulations by enabling the auditor to query the model in an oblivious manner. Our approach leverages a Private Information Retrieval mechanism to require the provider to label a large set of instances, while preventing it from knowing which subset will ultimately be used for the audit. The protocol is efficient, imposes minimal overhead on the auditor, and requires no modification to the audited model, its training procedure, or its inference pipeline. We provide theoretical guarantees showing that, under this protocol, a provider attempting to hide unfairness must falsify a significantly larger number of responses, thereby increasing both the difficulty and the likelihood of detection of manipulation. Experimental results across representative audit scenarios confirm the effectiveness and practicality of our approach.
