Affine Variational Autoencoders: An Efficient Approach for Improving Generalization and Robustness to Distribution Shift
This work addresses robustness to distribution shift for VAE users, but it is incremental as it focuses specifically on affine perturbations.
The paper tackles the problem of Variational Autoencoders (VAEs) failing to generalize to affine perturbations by proposing the Affine Variational Autoencoder (AVAE), which optimizes an affine transform to maximize ELBO and improves robustness without increasing model complexity, with experiments showing significant gains in generalization and robustness to distributional shift.
In this study, we propose the Affine Variational Autoencoder (AVAE), a variant of Variational Autoencoder (VAE) designed to improve robustness by overcoming the inability of VAEs to generalize to distributional shifts in the form of affine perturbations. By optimizing an affine transform to maximize ELBO, the proposed AVAE transforms an input to the training distribution without the need to increase model complexity to model the full distribution of affine transforms. In addition, we introduce a training procedure to create an efficient model by learning a subset of the training distribution, and using the AVAE to improve generalization and robustness to distributional shift at test time. Experiments on affine perturbations demonstrate that the proposed AVAE significantly improves generalization and robustness to distributional shift in the form of affine perturbations without an increase in model complexity.