Abstract
The focus of disentanglement approaches has been on identifying independent factors of variation in data. However, the causal variables underlying real-world observations are often not statistically independent. In this work, we bridge the gap to real-world scenarios by analyzing the behavior of the most prominent disentanglement approaches on correlated data in a large-scale empirical study (including 4260 models). We show and quantify that systematically induced correlations in the dataset are being learned and reflected in the latent representations, which has implications for downstream applications of disentanglement such as fairness. We also demonstrate how to resolve these latent correlations, either using weak supervision during training or by post-hoc correcting a
pre-trained model with a small number of labels.
pre-trained model with a small number of labels.
Original language | English |
---|---|
Title of host publication | Proceedings of the 38th International Conference on Machine Learning |
Number of pages | 12 |
Publisher | International Machine Learning Society (IMLS) |
Publication date | 2021 |
Publication status | Published - 2021 |
Event | 38th International Conference on Machine Learning - Virtual event Duration: 18 Jul 2021 → 24 Jul 2021 Conference number: 38 https://icml.cc/Conferences/2021 |
Conference
Conference | 38th International Conference on Machine Learning |
---|---|
Number | 38 |
Location | Virtual event |
Period | 18/07/2021 → 24/07/2021 |
Internet address |
Series | Proceedings of Machine Learning Research |
---|---|
Volume | 139 |
ISSN | 2640-3498 |