Forget-me-not! Contrastive Critics for Mitigating Posterior Collapse
This work addresses a key limitation in VAEs for researchers in probabilistic modeling, offering a faster and effective solution that bridges contrastive learning and variational inference.
The paper tackles posterior collapse in variational autoencoders by introducing inference critics that enforce correspondence between latent variables and observations, increasing mutual information and achieving competitive results on three datasets.
Variational autoencoders (VAEs) suffer from posterior collapse, where the powerful neural networks used for modeling and inference optimize the objective without meaningfully using the latent representation. We introduce inference critics that detect and incentivize against posterior collapse by requiring correspondence between latent variables and the observations. By connecting the critic's objective to the literature in self-supervised contrastive representation learning, we show both theoretically and empirically that optimizing inference critics increases the mutual information between observations and latents, mitigating posterior collapse. This approach is straightforward to implement and requires significantly less training time than prior methods, yet obtains competitive results on three established datasets. Overall, the approach lays the foundation to bridge the previously disconnected frameworks of contrastive learning and probabilistic modeling with variational autoencoders, underscoring the benefits both communities may find at their intersection.