Skip to main content Skip to main navigation


Hierarchical Decomposition of Nonlinear Dynamics and Control for System Identification and Policy Distillation

Hany Abdulsamad; Jan Peters
In: Alexandre M. Bayen; Ali Jadbabaie; George J. Pappas; Pablo A. Parrilo; Benjamin Recht; Claire J. Tomlin; Melanie N. Zeilinger (Hrsg.). Proceedings of the 2nd Annual Conference on Learning for Dynamics and Control. Learning for Dynamics and Control Conference (L4DC-2020), June 11-12, Berkeley, CA, USA, Pages 904-914, Proceedings of Machine Learning Research, Vol. 120, PMLR, 2020.


The control of nonlinear dynamical systems remains a major challenge for autonomous agents. Current trends in reinforcement learning (RL) focus on complex representations of dynamics and policies, which have yielded impressive results in solving a variety of hard control tasks. However, this new sophistication and extremely over-parameterized models have come with the cost of an overall reduction in our ability to interpret the resulting policies. In this paper, we take inspiration from the control community and apply the principles of hybrid switching systems in order to break down complex dynamics into simpler components. We exploit the rich representational power of probabilistic graphical models and derive an expectation-maximization (EM) algorithm for learning a sequence model to capture the temporal structure of the data and automatically decompose nonlinear dynamics into stochastic switching linear dynamical systems. Moreover, we show how this framework of switching models enables extracting hierarchies of Markovian and auto-regressive locally linear controllers from nonlinear experts in an imitation learning scenario.

Weitere Links