Faire: Repairing fairness of neural networks via neuron condition synthesis

Deep Neural Networks (DNNs) have achieved tremendous success in many applications, while it has been demonstrated that DNNs can exhibit some undesirable behaviors on concerns such as robustness, privacy, and other trustworthiness issues. Among them, fairness (i.e., non-discrimination) is one importa...

Full description

Saved in:
Bibliographic Details
Main Authors: LI, Tianlin, XIE, Xiaofei, WANG, Jian, GUO, Qing, LIU, Aishan, MA, Lei, LIU, Yang
Format: text
Language:English
Published: Institutional Knowledge at Singapore Management University 2023
Subjects:
Online Access:https://ink.library.smu.edu.sg/sis_research/8475
https://ink.library.smu.edu.sg/context/sis_research/article/9478/viewcontent/Faire__Repairing_Fairness_of_Neural_Networks_via_Neuron_Condition_Synthesis.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Singapore Management University
Language: English
id sg-smu-ink.sis_research-9478
record_format dspace
spelling sg-smu-ink.sis_research-94782024-01-04T09:12:42Z Faire: Repairing fairness of neural networks via neuron condition synthesis LI, Tianlin XIE, Xiaofei WANG, Jian GUO, Qing LIU, Aishan MA, Lei LIU, Yang Deep Neural Networks (DNNs) have achieved tremendous success in many applications, while it has been demonstrated that DNNs can exhibit some undesirable behaviors on concerns such as robustness, privacy, and other trustworthiness issues. Among them, fairness (i.e., non-discrimination) is one important property, especially when they are applied to some sensitive applications (e.g., finance and employment). However, DNNs easily learn spurious correlations between protected attributes (e.g., age, gender, race) and the classification task and develop discriminatory behaviors if the training data is imbalanced. Such discriminatory decisions in sensitive applications would introduce severe social impacts. To expose potential discrimination problems in DNNs before putting them in use, some testing techniques have been proposed to identify the discriminatory instances (i.e., instances that show defined discrimination1). However, how to repair DNNs after detecting such discrimination is still challenging. Existing techniques mainly rely on retraining on a large number of discriminatory instances generated by testing methods, which requires huge time overhead and makes the repairing inefficient.In this work, we propose the method Faire to effectively and efficiently repair the fairness issues of DNNs, without using additional data (e.g., discriminatory instances). Our basic idea is inspired by the traditional program repair method that synthesizes proper condition checking. To repair traditional programs, a typical method is to localize the program defects and repair the program logic by adding condition checking. Similarly, for DNNs, we try to understand the unfair logic and reformulate it with well-designed condition checking. In this article, we synthesize the condition that can reduce the effect of features relevant to the protected attributes in the DNN. Specifically, we first perform the neuron-based analysis and check the functionalities of neurons to identify neurons whose outputs could be regarded as features relevant to protected attributes and original tasks. Then a new condition layer is added after each hidden layer to penalize neurons that are accountable for the protected features (i.e., intermediate features relevant to protected attributes) and promote neurons that are accountable for the non-protected features (i.e., intermediate features relevant to original tasks). In sum, the repair rate2 of Faire reaches up to more than 99%, which outperforms other methods, and the whole repairing process only takes no more than 340 s. The evaluation results demonstrate that our approach can effectively and efficiently repair the individual discriminatory instances of the target model. 2023-11-01T07:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/8475 info:doi/10.1145/3617168 https://ink.library.smu.edu.sg/context/sis_research/article/9478/viewcontent/Faire__Repairing_Fairness_of_Neural_Networks_via_Neuron_Condition_Synthesis.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University Computing methodologies Machine learning Machine learning approaches Neural networks Software and its engineering Software creation and management Software verification and validation Software defect analysis Software testing and debugging Databases and Information Systems
institution Singapore Management University
building SMU Libraries
continent Asia
country Singapore
Singapore
content_provider SMU Libraries
collection InK@SMU
language English
topic Computing methodologies
Machine learning
Machine learning approaches
Neural networks
Software and its engineering
Software creation and management
Software verification and validation
Software defect analysis
Software testing and debugging
Databases and Information Systems
spellingShingle Computing methodologies
Machine learning
Machine learning approaches
Neural networks
Software and its engineering
Software creation and management
Software verification and validation
Software defect analysis
Software testing and debugging
Databases and Information Systems
LI, Tianlin
XIE, Xiaofei
WANG, Jian
GUO, Qing
LIU, Aishan
MA, Lei
LIU, Yang
Faire: Repairing fairness of neural networks via neuron condition synthesis
description Deep Neural Networks (DNNs) have achieved tremendous success in many applications, while it has been demonstrated that DNNs can exhibit some undesirable behaviors on concerns such as robustness, privacy, and other trustworthiness issues. Among them, fairness (i.e., non-discrimination) is one important property, especially when they are applied to some sensitive applications (e.g., finance and employment). However, DNNs easily learn spurious correlations between protected attributes (e.g., age, gender, race) and the classification task and develop discriminatory behaviors if the training data is imbalanced. Such discriminatory decisions in sensitive applications would introduce severe social impacts. To expose potential discrimination problems in DNNs before putting them in use, some testing techniques have been proposed to identify the discriminatory instances (i.e., instances that show defined discrimination1). However, how to repair DNNs after detecting such discrimination is still challenging. Existing techniques mainly rely on retraining on a large number of discriminatory instances generated by testing methods, which requires huge time overhead and makes the repairing inefficient.In this work, we propose the method Faire to effectively and efficiently repair the fairness issues of DNNs, without using additional data (e.g., discriminatory instances). Our basic idea is inspired by the traditional program repair method that synthesizes proper condition checking. To repair traditional programs, a typical method is to localize the program defects and repair the program logic by adding condition checking. Similarly, for DNNs, we try to understand the unfair logic and reformulate it with well-designed condition checking. In this article, we synthesize the condition that can reduce the effect of features relevant to the protected attributes in the DNN. Specifically, we first perform the neuron-based analysis and check the functionalities of neurons to identify neurons whose outputs could be regarded as features relevant to protected attributes and original tasks. Then a new condition layer is added after each hidden layer to penalize neurons that are accountable for the protected features (i.e., intermediate features relevant to protected attributes) and promote neurons that are accountable for the non-protected features (i.e., intermediate features relevant to original tasks). In sum, the repair rate2 of Faire reaches up to more than 99%, which outperforms other methods, and the whole repairing process only takes no more than 340 s. The evaluation results demonstrate that our approach can effectively and efficiently repair the individual discriminatory instances of the target model.
format text
author LI, Tianlin
XIE, Xiaofei
WANG, Jian
GUO, Qing
LIU, Aishan
MA, Lei
LIU, Yang
author_facet LI, Tianlin
XIE, Xiaofei
WANG, Jian
GUO, Qing
LIU, Aishan
MA, Lei
LIU, Yang
author_sort LI, Tianlin
title Faire: Repairing fairness of neural networks via neuron condition synthesis
title_short Faire: Repairing fairness of neural networks via neuron condition synthesis
title_full Faire: Repairing fairness of neural networks via neuron condition synthesis
title_fullStr Faire: Repairing fairness of neural networks via neuron condition synthesis
title_full_unstemmed Faire: Repairing fairness of neural networks via neuron condition synthesis
title_sort faire: repairing fairness of neural networks via neuron condition synthesis
publisher Institutional Knowledge at Singapore Management University
publishDate 2023
url https://ink.library.smu.edu.sg/sis_research/8475
https://ink.library.smu.edu.sg/context/sis_research/article/9478/viewcontent/Faire__Repairing_Fairness_of_Neural_Networks_via_Neuron_Condition_Synthesis.pdf
_version_ 1787590776363941888