Timezone: »

Contributed talk: Safe Policy Search with Gaussian Process Models
Kyriakos Polymenakos · Stephen J Roberts

We propose a method to optimise the parameters of a policy which will be used to safely perform a given task in a data-efficient manner. We train a Gaussian process model to capture the system dynamics, based on the PILCO framework. Our model has useful analytic properties, which allow closed form computation of error gradients and estimating the probability of violating given state space constraints. During training, as well as operation, only policies that are deemed safe are implemented on the real system, minimising the risk of failure.

Author Information

Kyriakos Polymenakos (Oxford University - Machine Learning Research Group)

Phd student in Machine Learning at Oxford University. My research is focused on safe model-based policy search methods. Interests include reinforcement learning, deep learning, safety and verification, control theory.

Stephen J Roberts (University of Oxford)

More from the Same Authors