GEM: Group Enhanced Model for Learning Dynamical Control Systems

by   Philippe Hansen-Estruch, et al.

Learning the dynamics of a physical system wherein an autonomous agent operates is an important task. Often these systems present apparent geometric structures. For instance, the trajectories of a robotic manipulator can be broken down into a collection of its transitional and rotational motions, fully characterized by the corresponding Lie groups and Lie algebras. In this work, we take advantage of these structures to build effective dynamical models that are amenable to sample-based learning. We hypothesize that learning the dynamics on a Lie algebra vector space is more effective than learning a direct state transition model. To verify this hypothesis, we introduce the Group Enhanced Model (GEM). GEMs significantly outperform conventional transition models on tasks of long-term prediction, planning, and model-based reinforcement learning across a diverse suite of standard continuous-control environments, including Walker, Hopper, Reacher, Half-Cheetah, Inverted Pendulums, Ant, and Humanoid. Furthermore, plugging GEM into existing state of the art systems enhances their performance, which we demonstrate on the PETS system. This work sheds light on a connection between learning of dynamics and Lie group properties, which opens doors for new research directions and practical applications along this direction. Our code is publicly available at:


page 6

page 7

page 8


Geometric sliding mode control of mechanical systems on Lie groups

This paper presents a generalization of conventional sliding mode contro...

Lie Algebrized Gaussians for Image Representation

We present an image representation method which is derived from analyzin...

From Pixels to Torques: Policy Learning with Deep Dynamical Models

Data-efficient learning in continuous state-action spaces using very hig...

Lie Group Forced Variational Integrator Networks for Learning and Control of Robot Systems

Incorporating prior knowledge of physics laws and structural properties ...

Commutative Lie Group VAE for Disentanglement Learning

We view disentanglement learning as discovering an underlying structure ...

Learning Dynamics Models for Model Predictive Agents

Model-Based Reinforcement Learning involves learning a dynamics model fr...

Dynamics of the N-fold Pendulum in the framework of Lie Group Integrators

Since their introduction, Lie group integrators have become a method of ...

Please sign up or login with your details

Forgot password? Click here to reset