Fairness by Learning Orthogonal Disentangled Representations
Sarhan MH, Navab N, Eslami A, Albarqouni S (2020)
Publication Type: Conference contribution
Publication year: 2020
Journal
Publisher: Springer Science and Business Media Deutschland GmbH
Book Volume: 12374 LNCS
Pages Range: 746-761
Conference Proceedings Title: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Event location: Glasgow, GBR
ISBN: 9783030585259
DOI: 10.1007/978-3-030-58526-6_44
Abstract
Learning discriminative powerful representations is a crucial step for machine learning systems. Introducing invariance against arbitrary nuisance or sensitive attributes while performing well on specific tasks is an important problem in representation learning. This is mostly approached by purging the sensitive information from learned representations. In this paper, we propose a novel disentanglement approach to invariant representation problem. We disentangle the meaningful and sensitive representations by enforcing orthogonality constraints as a proxy for independence. We explicitly enforce the meaningful representation to be agnostic to sensitive information by entropy maximization. The proposed approach is evaluated on five publicly available datasets and compared with state of the art methods for learning fairness and invariance achieving the state of the art performance on three datasets and comparable performance on the rest. Further, we perform an ablative study to evaluate the effect of each component.
Involved external institutions
Germany (DE)How to cite
APA:
Sarhan, M.H., Navab, N., Eslami, A., & Albarqouni, S. (2020). Fairness by Learning Orthogonal Disentangled Representations. In Andrea Vedaldi, Horst Bischof, Thomas Brox, Jan-Michael Frahm (Eds.), Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (pp. 746-761). Glasgow, GBR: Springer Science and Business Media Deutschland GmbH.
MLA:
Sarhan, Mhd Hasan, et al. "Fairness by Learning Orthogonal Disentangled Representations." Proceedings of the 16th European Conference on Computer Vision, ECCV 2020, Glasgow, GBR Ed. Andrea Vedaldi, Horst Bischof, Thomas Brox, Jan-Michael Frahm, Springer Science and Business Media Deutschland GmbH, 2020. 746-761.
BibTeX: Download