Computer Science and Engineering

Christian R. Shelton, Professor

Importance Sampling for Reinforcement Learning with Multiple Objectives (2001)

by Christian Robert Shelton

Abstract: This thesis considers three complications that arise from applying reinforcement learning to a real-world application. In the process of using reinforcement learning to build an adaptive electronic market-maker, we find the sparsity of data, the partial observability of the domain, and the multiple objectives of the agent to cause serious problems for existing reinforcement learning algorithms.

We employ importance sampling (likelihood ratios) to achieve good performance in partially observable Markov decision processes with few data. Our importance sampling estimator requires no knowledge about the environment and places few restrictions on the method of collecting data. It can be used efficiently with reactive controllers, finite-state controllers, or policies with function approximation. We present theoretical analyses of the estimator and incorporate it into a reinforcement learning algorithm.

Additionally, this method provides a complete return surface which can be used to balance multiple objectives dynamically. We demonstrate the need for multiple goals in a variety of applications and natural solutions based on our sampling method. The thesis concludes with example results from employing our algorithm to the domain of automated electronic market-making.

Download Information

Christian Robert Shelton (2001). Importance Sampling for Reinforcement Learning with Multiple Objectives. Doctoral dissertation, Massachusetts Institute of Technology. pdf   ps ps.gz    

Bibtex citation

   author = "Christian Robert Shelton",
   title = "Importance Sampling for Reinforcement Learning with Multiple Objectives",
   school = "Massachusetts Institute of Technology",
   schoolabbr = "{MIT}",
   year = 2001,
   month = Aug,

More Information


University of California, Riverside
Chung Hall, room 327
Riverside, CA 92521
Tel: (951) 827-2554
E-mail: cshelton@cs.ucr.edu


Other Links