Probabilistic Autoencoder using Fisher Information
This work addresses uncertainty quantification for researchers using autoencoders in fields like physics, though it appears incremental as an extension to existing VAE architectures.
The authors tackled the problem of uncertainty quantification in autoencoders by introducing FisherNet, which derives latent space uncertainty from the decoder using the Fisher information metric instead of an additional encoder channel. They showed experimentally that FisherNet produces more accurate data reconstructions than a comparable VAE and scales better with latent dimensions.
Neural Networks play a growing role in many science disciplines, including physics. Variational Autoencoders (VAEs) are neural networks that are able to represent the essential information of a high dimensional data set in a low dimensional latent space, which have a probabilistic interpretation. In particular the so-called encoder network, the first part of the VAE, which maps its input onto a position in latent space, additionally provides uncertainty information in terms of a variance around this position. In this work, an extension to the Autoencoder architecture is introduced, the FisherNet. In this architecture, the latent space uncertainty is not generated using an additional information channel in the encoder, but derived from the decoder, by means of the Fisher information metric. This architecture has advantages from a theoretical point of view as it provides a direct uncertainty quantification derived from the model, and also accounts for uncertainty cross-correlations. We can show experimentally that the FisherNet produces more accurate data reconstructions than a comparable VAE and its learning performance also apparently scales better with the number of latent space dimensions.