Policy Search in Reproducing Kernel Hilbert Space

J. Andrew (Drew) Bagnell and Jeff Schneider
tech. report CMU-RI-TR-03-45, Robotics Institute, Carnegie Mellon University, November, 2003

  • Adobe portable document format (pdf) (351KB)
Copyright notice: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder.

Much recent work in reinforcement learning and stochastic optimal control has focused on algorithms that search directly through a space of policies rather than building approximate value functions. Policy search has numerous advantages: it does not rely on the Markov assumption, domain knowledge may be encoded in a policy, the policy may require less representational power than a value-function approximation, and stable and convergent algorithms are well-understood. In contrast with value-function methods, however, existing approaches to policy search have heretofore focused entirely on parametric approaches. This places fundamental limits on the kind of policies that can be represented. In this work, we show how policy search (with or without the additional guidance of value-functions) in a Reproducing Kernel Hilbert Space gives a simple and rigorous extension of the technique to non-parametric settings.

In particular, we investigate a new class of algorithms which generalize \textsc{Reinforce}-style likelihood ratio methods to yield both online and batch techniques that perform gradient search in a function space of policies. Further, we describe the computational tools that allow efficient implementation. Finally, we apply our new techniques towards interesting reinforcement learning problems.

reinforcement learning, control, Reinforce, kernel methods, hilbert space

Associated Lab(s) / Group(s): Reliable Autonomous Systems Lab
Associated Project(s): Federation of Intelligent Robotic Explorers Project

Text Reference
J. Andrew (Drew) Bagnell and Jeff Schneider, "Policy Search in Reproducing Kernel Hilbert Space," tech. report CMU-RI-TR-03-45, Robotics Institute, Carnegie Mellon University, November, 2003

BibTeX Reference
   author = "J. Andrew (Drew) Bagnell and Jeff Schneider",
   title = "Policy Search in Reproducing Kernel Hilbert Space",
   booktitle = "",
   institution = "Robotics Institute",
   month = "November",
   year = "2003",
   number= "CMU-RI-TR-03-45",
   address= "Pittsburgh, PA",