Variational Temporal Abstraction

Abstract

We introduce a variational approach to learning and inference of temporally hierarchical structure and representation for sequential data. We propose the Variational Temporal Abstraction (VTA), a hierarchical recurrent state space model that can infer the latent temporal structure and thus perform the stochastic state transition hierarchically. We also propose to apply this model to implement the jumpy-imagination ability in imagination-augmented agent-learning in order to improve the efficiency of the imagination. In experiments, we demonstrate that our proposed method can model 2D and 3D visual sequence datasets with interpretable temporal structure discovery and that its application to jumpy imagination enables more efficient agent-learning in a 3D navigation task.

Accepted in NeurIPS 2019

Cited by in corpus (7)

• Stochastic Latent Actor-Critic: Deep Reinforcement Learning with a Latent Variable Model
• Long-Horizon Visual Planning with Goal-Conditioned Hierarchical Predictors
• Clockwork Variational Autoencoders
• Deep Explicit Duration Switching Models for Time Series
• Variational Predictive Routing with Nested Subjective Timescales
• A Temporal Variational Model for Story Generation
• Subgoal Search For Complex Reasoning Tasks