Weakly-supervised disentanglement without compromises

Locatello F, Poole B, Rätsch G, Schölkopf B, Bachem O, Tschannen M. 2020. Weakly-supervised disentanglement without compromises. Proceedings of the 37th International Conference on Machine Learning. International Conference on Machine Learning, PMLR, vol. 119, 6348–6359.

Download (ext.)
Conference Paper | Published | English

Scopus indexed
Author
Locatello, FrancescoISTA ; Poole, Ben; Rätsch, Gunnar; Schölkopf, Bernhard; Bachem, Olivier; Tschannen, Michael
Department
Series Title
PMLR
Abstract
Intelligent agents should be able to learn useful representations by observing changes in their environment. We model such observations as pairs of non-i.i.d. images sharing at least one of the underlying factors of variation. First, we theoretically show that only knowing how many factors have changed, but not which ones, is sufficient to learn disentangled representations. Second, we provide practical algorithms that learn disentangled representations from pairs of images without requiring annotation of groups, individual factors, or the number of factors that have changed. Third, we perform a large-scale empirical study and show that such pairs of observations are sufficient to reliably learn disentangled representations on several benchmark data sets. Finally, we evaluate our learned representations and find that they are simultaneously useful on a diverse suite of tasks, including generalization under covariate shifts, fairness, and abstract reasoning. Overall, our results demonstrate that weak supervision enables learning of useful disentangled representations in realistic scenarios.
Publishing Year
Date Published
2020-07-07
Proceedings Title
Proceedings of the 37th International Conference on Machine Learning
Volume
119
Page
6348–6359
Conference
International Conference on Machine Learning
Conference Location
Virtual
Conference Date
2020-07-13 – 2020-07-18
IST-REx-ID

Cite this

Locatello F, Poole B, Rätsch G, Schölkopf B, Bachem O, Tschannen M. Weakly-supervised disentanglement without compromises. In: Proceedings of the 37th International Conference on Machine Learning. Vol 119. ; 2020:6348–6359.
Locatello, F., Poole, B., Rätsch, G., Schölkopf, B., Bachem, O., & Tschannen, M. (2020). Weakly-supervised disentanglement without compromises. In Proceedings of the 37th International Conference on Machine Learning (Vol. 119, pp. 6348–6359). Virtual.
Locatello, Francesco, Ben Poole, Gunnar Rätsch, Bernhard Schölkopf, Olivier Bachem, and Michael Tschannen. “Weakly-Supervised Disentanglement without Compromises.” In Proceedings of the 37th International Conference on Machine Learning, 119:6348–6359, 2020.
F. Locatello, B. Poole, G. Rätsch, B. Schölkopf, O. Bachem, and M. Tschannen, “Weakly-supervised disentanglement without compromises,” in Proceedings of the 37th International Conference on Machine Learning, Virtual, 2020, vol. 119, pp. 6348–6359.
Locatello F, Poole B, Rätsch G, Schölkopf B, Bachem O, Tschannen M. 2020. Weakly-supervised disentanglement without compromises. Proceedings of the 37th International Conference on Machine Learning. International Conference on Machine Learning, PMLR, vol. 119, 6348–6359.
Locatello, Francesco, et al. “Weakly-Supervised Disentanglement without Compromises.” Proceedings of the 37th International Conference on Machine Learning, vol. 119, 2020, pp. 6348–6359.
All files available under the following license(s):
Copyright Statement:
This Item is protected by copyright and/or related rights. [...]

Link(s) to Main File(s)
Access Level
OA Open Access

Export

Marked Publications

Open Data ISTA Research Explorer

Sources

arXiv 2002.02886

Search this title in

Google Scholar