FAVAE: Sequence Disentanglement using Information Bottleneck Principle
This addresses the challenge of interpretable and transferable representation learning for sequential data such as video and speech, though it appears incremental as it builds on prior disentanglement methods.
The authors tackled the problem of learning disentangled representations from sequential data by proposing FAVAE, a generative model that disentangles multiple dynamic factors without supervision, achieving state-of-the-art results in extracting interpretable factors like 'picking up' and 'throwing' in robotic tasks.
We propose the factorized action variational autoencoder (FAVAE), a state-of-the-art generative model for learning disentangled and interpretable representations from sequential data via the information bottleneck without supervision. The purpose of disentangled representation learning is to obtain interpretable and transferable representations from data. We focused on the disentangled representation of sequential data since there is a wide range of potential applications if disentanglement representation is extended to sequential data such as video, speech, and stock market. Sequential data are characterized by dynamic and static factors: dynamic factors are time dependent, and static factors are independent of time. Previous models disentangle static and dynamic factors by explicitly modeling the priors of latent variables to distinguish between these factors. However, these models cannot disentangle representations between dynamic factors, such as disentangling "picking up" and "throwing" in robotic tasks. FAVAE can disentangle multiple dynamic factors. Since it does not require modeling priors, it can disentangle "between" dynamic factors. We conducted experiments to show that FAVAE can extract disentangled dynamic factors.