MAFT: Efficient Model-Agnostic Fairness Testing for Deep Neural Networks via Zero-Order Gradient Search
This addresses fairness concerns in decision-making systems using DNNs by providing a scalable black-box testing method, though it is incremental as it builds on existing fairness testing paradigms.
The paper tackles the problem of black-box individual fairness testing for deep neural networks, proposing MAFT, which matches state-of-the-art white-box methods in effectiveness while improving applicability to large-scale networks, with performance gains of approximately 14.69 times in effectiveness and 32.58 times in efficiency over existing black-box approaches.
Deep neural networks (DNNs) have shown powerful performance in various applications and are increasingly being used in decision-making systems. However, concerns about fairness in DNNs always persist. Some efficient white-box fairness testing methods about individual fairness have been proposed. Nevertheless, the development of black-box methods has stagnated, and the performance of existing methods is far behind that of white-box methods. In this paper, we propose a novel black-box individual fairness testing method called Model-Agnostic Fairness Testing (MAFT). By leveraging MAFT, practitioners can effectively identify and address discrimination in DL models, regardless of the specific algorithm or architecture employed. Our approach adopts lightweight procedures such as gradient estimation and attribute perturbation rather than non-trivial procedures like symbol execution, rendering it significantly more scalable and applicable than existing methods. We demonstrate that MAFT achieves the same effectiveness as state-of-the-art white-box methods whilst improving the applicability to large-scale networks. Compared to existing black-box approaches, our approach demonstrates distinguished performance in discovering fairness violations w.r.t effectiveness (approximately 14.69 times) and efficiency (approximately 32.58 times).