paper-with-me

홈 › Papers

On Disentangled Representations Learned From Correlated Data

2020-06-14 · Frederik Träuble, Elliot Creager, Niki Kilbertus, Francesco Locatello, Andrea Dittadi, Anirudh Goyal, Bernhard Schölkopf, Stefan Bauer

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.

📄 PDF Abstract BibTeX arXiv:2006.07886

Code (2)

ftraeuble/disentanglement_lib 공식 구현 tf
facebookresearch/disentangling-correlated-factors pytorch

Tasks

DisentanglementFairness

Similar Papers 제목 키워드 기반

On Disentangled and Locally Fair Representations

2022-05-05 · Yaron Gurovich, Sagie Benaim, Lior Wolf

We study the problem of performing classification in a manner that is fair for sensitive groups, such as race and gender. This problem is tackled through the lens of disentangled and locally fair representations. We lear…

AttributeFairness

Be More Active! Understanding the Differences between Mean and Sampled Representations of Variational Autoencoders

2021-09-26 · Lisa Bonheme, Marek Grzes

The ability of Variational Autoencoders to learn disentangled representations has made them appealing for practical applications. However, their mean representations, which are generally used for downstream tasks, have r…

Disentanglement

On the Fairness of Disentangled Representations

2019-05-31 · NeurIPS 2019 12 · Francesco Locatello, Gabriele Abbati, Tom Rainforth, Stefan Bauer 외

Recently there has been a significant interest in learning disentangled representations, as they promise increased interpretability, generalization to unseen scenarios and faster learning on downstream tasks. In this pap…

DisentanglementFairness

Conditional Mutual Information for Disentangled Representations in Reinforcement Learning

2023-05-23 · NeurIPS 2023 11 · Mhairi Dunion, Trevor McInroe, Kevin Sebastian Luck, Josiah P. Hanna 외

Reinforcement Learning (RL) environments can produce training data with spurious correlations between features due to the amount of training data or its limited feature coverage. This can lead to RL agents encoding these…

continuous-controlContinuous ControlDisentanglementreinforcement-learning+2

Scaling-up Disentanglement for Image Translation

2021-03-25 · ICCV 2021 10 · Aviv Gabbay, Yedid Hoshen

Image translation methods typically aim to manipulate a set of labeled attributes (given as supervision at training time e.g. domain label) while leaving the unlabeled attributes intact. Current methods achieve either: (…

DisentanglementDiversityTranslation