Disentangling Dynamics and Content for Control and Planning
This addresses the challenge of leveraging limited action-labeled data for control in dynamical systems, which is incremental as it builds on existing representation learning methods.
The paper tackles the problem of learning a controllable representation from high-dimensional observations of dynamical systems, where only one dataset includes action effects, and achieves a representation usable for planning and long-term prediction.
In this paper, We study the problem of learning a controllable representation for high-dimensional observations of dynamical systems. Specifically, we consider a situation where there are multiple sets of observations of dynamical systems with identical underlying dynamics. Only one of these sets has information about the effect of actions on the observation and the rest are just some random observations of the system. Our goal is to utilize the information in that one set and find a representation for the other sets that can be used for planning and ling-term prediction.