Variational Temporal Abstraction
This work addresses the need for efficient temporal abstraction in sequential modeling and reinforcement learning, though it appears incremental as it builds on existing hierarchical and variational methods.
The paper tackled the problem of learning hierarchical temporal structure in sequential data and improving agent-learning efficiency, resulting in interpretable temporal structure discovery on 2D/3D visual sequences and more efficient agent-learning in a 3D navigation task.
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.