TY - GEN
T1 - Fast and Efficient MMD-Based Fair PCA via Optimization over Stiefel Manifold
AU - Lee, Junghyun
AU - Kim, Gwangsu
AU - Olfat, Mahbod
AU - Hasegawa-Johnson, Mark
AU - Yoo, Chang D.
N1 - Publisher Copyright:
© 2022, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.
PY - 2022/6/30
Y1 - 2022/6/30
N2 - This paper defines fair principal component analysis (PCA) as minimizing the maximum mean discrepancy (MMD) between dimensionality-reduced conditional distributions of different protected classes. The incorporation of MMD naturally leads to an exact and tractable mathematical formulation of fairness with good statistical properties. We formulate the problem of fair PCA subject to MMD constraints as a non-convex optimization over the Stiefel manifold and solve it using the Riemannian Exact Penalty Method with Smoothing (REPMS). Importantly, we provide local optimality guarantees and explicitly show the theoretical effect of each hyperparameter in practical settings, extending previous results. Experimental comparisons based on synthetic and UCI datasets show that our approach outperforms prior work in explained variance, fairness, and runtime.
AB - This paper defines fair principal component analysis (PCA) as minimizing the maximum mean discrepancy (MMD) between dimensionality-reduced conditional distributions of different protected classes. The incorporation of MMD naturally leads to an exact and tractable mathematical formulation of fairness with good statistical properties. We formulate the problem of fair PCA subject to MMD constraints as a non-convex optimization over the Stiefel manifold and solve it using the Riemannian Exact Penalty Method with Smoothing (REPMS). Importantly, we provide local optimality guarantees and explicitly show the theoretical effect of each hyperparameter in practical settings, extending previous results. Experimental comparisons based on synthetic and UCI datasets show that our approach outperforms prior work in explained variance, fairness, and runtime.
UR - http://www.scopus.com/inward/record.url?scp=85139279109&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85139279109&partnerID=8YFLogxK
U2 - 10.1609/aaai.v36i7.20699
DO - 10.1609/aaai.v36i7.20699
M3 - Conference contribution
AN - SCOPUS:85139279109
T3 - Proceedings of the 36th AAAI Conference on Artificial Intelligence, AAAI 2022
SP - 7363
EP - 7371
BT - AAAI-22 Technical Tracks 7
PB - Association for the Advancement of Artificial Intelligence
T2 - 36th AAAI Conference on Artificial Intelligence, AAAI 2022
Y2 - 22 February 2022 through 1 March 2022
ER -