TY - GEN
T1 - Fair Representation Learning
T2 - 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, KDD 2022
AU - Liu, Ji
AU - Li, Zenan
AU - Yao, Yuan
AU - Xu, Feng
AU - Ma, Xiaoxing
AU - Xu, Miao
AU - Tong, Hanghang
N1 - This work is supported by the National Natural Science Foundation of China (No. 62025202), the Collaborative Innovation Center of Novel Software Technology and Industrialization, and the Fundamental Research Funds for the Central Universities. Hanghang Tong is partially supported by NSF (1947135, 2134079 and 1939725). Yuan Yao is the corresponding author.
PY - 2022/8/14
Y1 - 2022/8/14
N2 - Learning fair representations is an essential task to reduce bias in data-oriented decision making. It protects minority subgroups by requiring the learned representations to be independent of sensitive attributes. To achieve independence, the vast majority of the existing work primarily relaxes it to the minimization of the mutual information between sensitive attributes and learned representations. However, direct computation of mutual information is computationally intractable, and various upper bounds currently used either are still intractable or contradict the utility of the learned representations. In this paper, we introduce distance covariance as a new dependence measure into fair representation learning. By observing that sensitive attributes (e.g., gender, race, and age group) are typically categorical, the distance covariance can be converted to a tractable penalty term without contradicting the utility desideratum. Based on the tractable penalty, we propose FairDisCo, a variational method to learn fair representations. Experiments demonstrate that FairDisCo outperforms existing competitors for fair representation learning.
AB - Learning fair representations is an essential task to reduce bias in data-oriented decision making. It protects minority subgroups by requiring the learned representations to be independent of sensitive attributes. To achieve independence, the vast majority of the existing work primarily relaxes it to the minimization of the mutual information between sensitive attributes and learned representations. However, direct computation of mutual information is computationally intractable, and various upper bounds currently used either are still intractable or contradict the utility of the learned representations. In this paper, we introduce distance covariance as a new dependence measure into fair representation learning. By observing that sensitive attributes (e.g., gender, race, and age group) are typically categorical, the distance covariance can be converted to a tractable penalty term without contradicting the utility desideratum. Based on the tractable penalty, we propose FairDisCo, a variational method to learn fair representations. Experiments demonstrate that FairDisCo outperforms existing competitors for fair representation learning.
KW - distance covariance
KW - fair representation learning
KW - mutual information
UR - http://www.scopus.com/inward/record.url?scp=85137143244&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85137143244&partnerID=8YFLogxK
U2 - 10.1145/3534678.3539302
DO - 10.1145/3534678.3539302
M3 - Conference contribution
AN - SCOPUS:85137143244
T3 - Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
SP - 1088
EP - 1097
BT - KDD 2022 - Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining
PB - Association for Computing Machinery
Y2 - 14 August 2022 through 18 August 2022
ER -