RUNNER: Responsible UNfair NEuron Repair for Enhancing Deep Neural Network Fairness
Deep Neural Networks (DNNs), an emerging software technology, have achieved impressive results in a variety of fields. However, the discriminatory behaviors towards certain groups (a.k.a. unfairness) of DNN models increasingly become a social concern, especially in high-stake applications such as loan approval and criminal risk assessment. Although there has been a number of works to improve model fairness, most of them adopt an adversary to either expand the model architecture or augment training data, which introduces excessive computational overhead. Recent work diagnoses responsible unfair neurons first and fixes them with selective retraining. Unfortunately, existing diagnosis process is time-consuming due to multi-step training sample analysis and selective retraining may cause a performance bottleneck due to indirectly adjusting unfair neurons on biased samples. In this paper, we propose Responsible UNfair NEuron Repair (RUNNER) that improves existing works in three key aspects: (1) efficiency: we design the Importance-based Neuron Diagnosis that identifies responsible unfair neurons in one step with a novel importance criterion of neurons; (2) effectiveness: we design the Neuron Stabilizing Retraining by adding a loss term that measures the activation distance of responsible unfair neurons from different subgroups in all sources; (3) generalization: we investigate the effectiveness on both structured tabular data and large-scale unstructured image data, which is often ignored in prior studies. Our extensive experiments across 5 datasets show that RUUNER can effectively and efficiently diagnose and repair the DNNs regarding unfairness. On average, our approach significantly reduces computing overhead from 341.7s to 29.65s, and achieves improved fairness up to 79.3%. Besides, RUNNER also keeps state-of-the-art results on the unstructured dataset.
Wed 17 AprDisplayed time zone: Lisbon change
16:00 - 17:30 | |||
16:00 15mTalk | RUNNER: Responsible UNfair NEuron Repair for Enhancing Deep Neural Network Fairness Research Track Li Tianlin Nanyang Technological University, Yue Cao Nanyang Technological University, Jian Zhang Nanyang Technological University, Shiqian Zhao Nanyang Technological University, Yihao Huang East China Normal University, Aishan Liu Beihang University; Institute of Dataspace, Qing Guo IHPC and CFAR at A*STAR, Singapore, Yang Liu Nanyang Technological University | ||
16:15 15mTalk | ITER: Iterative Neural Repair for Multi-Location Patches Research Track | ||
16:30 15mTalk | Out of Context: How important is Local Context in Neural Program Repair? Research Track | ||
16:45 15mTalk | Out of Sight, Out of Mind: Better Automatic Vulnerability Repair by Broadening Input Ranges and Sources Research Track Xin Zhou Singapore Management University, Singapore, Kisub Kim Singapore Management University, Singapore, Bowen Xu North Carolina State University, DongGyun Han Royal Holloway, University of London, David Lo Singapore Management University | ||
17:00 15mTalk | Strengthening Supply Chain Security with Fine-grained Safe Patch Identification Research Track Luo Changhua The Chinese University of Hong Kong, Wei Meng Chinese University of Hong Kong, Shuai Wang The Hong Kong University of Science and Technology |