Skip to yearly menu bar Skip to main content


Poster
in
Workshop: Bayesian Deep Learning

Certifiably Robust Variational Autoencoders

Ben Barrett · Alexander Camuto · Matthew Willetts · Thomas Rainforth


Abstract:

We derive bounds on the minimal size of an input perturbation required to change a VAE’s reconstruction by more than an allowed amount, with these bounds depending on key parameters such as the Lipschitz constants of the encoder and decoder. Our bounds allow one to specify a desired level of robustness upfront and then train a VAE that is certified to achieve this robustness.

Chat is not available.