University of Cambridge > Talks.cam > CUED Control Group Seminars > Incorporating Domain-Specific Knowledge in Learning Control using Multiple Dynamics Models

Incorporating Domain-Specific Knowledge in Learning Control using Multiple Dynamics Models

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Tim Hughes.

We propose a probabilistic learning algorithm, based on that of Deisenroth & Rasmussen (2011), to train control policies for nonlinear systems with unknown, or partially unknown, dynamics. In particular, we address the issue of how to incorporate domain-specific knowledge in the form of known, or approximate, relationships between the state variables. This covers the common case of position-velocity relationships but can also be used to tackle reference tracking problems. We demonstrate our approach on some learning problems including the simulated robotic unicycle.

Deisenroth, M.P. & Rasmussen, C.E. (2011) PILCO : A model-based and data-efficient approach to policy search. In Proceedings of the 28th International Conference on Machine Learning (ICML)

This talk is part of the CUED Control Group Seminars series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2017 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity