Intelligent machines using machine learning algorithms are ubiquitous, ranging from simple
data analysis and pattern recognition tools to complex systems that achieve superhuman …

Not all generate-and-test search algorithms are created equal. Bayesian Optimization (BO)
invests a lot of computation time to generate the candidate solution that best balances the …

We introduce a method for constructing skills capable of solving tasks drawn from a
distribution of parameterized reinforcement learning problems. The method draws example …

In this paper, we present a robotic model-based reinforcement learning method that
combines ideas from model identification and model predictive control. We use a feature …

Tasks with unknown dynamics and costly system interaction time present a serious
challenge for reinforcement learning. If a model of the dynamics can be learned quickly …

Model predictive control (MPC) has been successful in applications involving the control of
complex physical systems. This class of controllers leverages the information provided by an …

Natural actor-critics are a popular class of policy search algorithms for finding locally optimal
policies for Markov decision processes. In this paper we address a drawback of natural actor …

We present new global and local policy search algorithms suitable for problems with policy-
dependent cost variance (or risk), a property present in many robot control tasks. These …

We introduce a method for actively learning parameterized skills. Parameterized skills are
flexible behaviors that can solve any task drawn from a distribution of parameterized …

Machine learning algorithms are everywhere, ranging from simple data analysis and pattern
recognition tools used across the sciences to complex systems that achieve super-human …