White-box fairness testing through adversarial sampling

View/ Open
File version
Accepted Manuscript (AM)
Author(s)
Zhang, P
Wang, J
Sun, J
Dong, G
Wang, X
Wang, X
Dong, JS
Dai, T
Griffith University Author(s)
Year published
2020
Metadata
Show full item recordAbstract
Although deep neural networks (DNNs) have demonstrated astonishing performance in many applications, there are still concerns on their dependability. One desirable property of DNN for applications with societal impact is fairness (i.e., non-discrimination). In this work, we propose a scalable approach for searching individual discriminatory instances of DNN. Compared with state-of-the-art methods, our approach only employs lightweight procedures like gradient computation and clustering, which makes it significantly more scalable than existing methods. Experimental results show that our approach explores the search space more ...
View more >Although deep neural networks (DNNs) have demonstrated astonishing performance in many applications, there are still concerns on their dependability. One desirable property of DNN for applications with societal impact is fairness (i.e., non-discrimination). In this work, we propose a scalable approach for searching individual discriminatory instances of DNN. Compared with state-of-the-art methods, our approach only employs lightweight procedures like gradient computation and clustering, which makes it significantly more scalable than existing methods. Experimental results show that our approach explores the search space more effectively (9 times) and generates much more individual discriminatory instances (25 times) using much less time (half to 1/7).
View less >
View more >Although deep neural networks (DNNs) have demonstrated astonishing performance in many applications, there are still concerns on their dependability. One desirable property of DNN for applications with societal impact is fairness (i.e., non-discrimination). In this work, we propose a scalable approach for searching individual discriminatory instances of DNN. Compared with state-of-the-art methods, our approach only employs lightweight procedures like gradient computation and clustering, which makes it significantly more scalable than existing methods. Experimental results show that our approach explores the search space more effectively (9 times) and generates much more individual discriminatory instances (25 times) using much less time (half to 1/7).
View less >
Conference Title
Proceedings 2020 ACM/IEEE 42nd International Conference on Software Engineering, ICSE 2020
Copyright Statement
© ACM, 2020. This is the author's version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version was published in ICSE '20: Proceedings of the ACM/IEEE 42nd International Conference on Software Engineering, ISBN: 978-1-4503-7121-6, https://doi.org/10.1145/3377811.3380331
Subject
Artificial intelligence