Group-Disentangling Conditional Shift

Dan Andrei Iliescu, Damon WischikICLR 2023 · OpenReview

Abstract

We propose a novel group disentanglement method called the Context-Aware Variational Autoencoder (CxVAE). Our model can learn disentangled representations on datasets with conditional shift. This phenomenon occurs when the conditional distribution of the instance-level latent variable $\mathbf{z}$ given the input observation $\mathbf{x}$ changes from one group to another (i.e. $p_i(\mathbf{z}|\mathbf{x}) \neq p_j(\mathbf{z}|\mathbf{x})$, where $i,j$ are two different groups). We show that existing methods fail to learn disentangled representations under this scenario because they infer the group $\mathbf{u}$ and instance $\mathbf{z}$ variables separately. CxVAE overcomes this limitation by conditioning the instance inference on the group variable $q(\mathbf{z}|\mathbf{x},\mathbf{u})$. Our model has the novel ability to disentangle ambiguous observations (those with incomplete information about the generative factors), which we evaluate on the task of fair comparisons between student test scores. Additionally, we demonstrate empirically that conditional shift is the cause of our model's improved performance.
BibTeX
@inproceedings{iliescu2023group,
  author = {Dan Andrei Iliescu and Damon Wischik},
  title = {{Group-Disentangling Conditional Shift}},
  booktitle = {International Conference on Learning Representations},
  year = {2023},
  url = {https://openreview.net/forum?id=d5U-bPKPde},
}

Loading related content...

Figure