Value Function Approximation in Reinforcement Learning Using the Fourier Basis

Authors

  • George Konidaris Massachusetts Institute of Technology
  • Sarah Osentoski Brown University
  • Philip Thomas University of Massachusetts Amherst

Abstract

We describe the Fourier basis, a linear value function approximation scheme based on the Fourier series. We empirically demonstrate that it performs well compared to radial basis functions and the polynomial basis, the two most popular fixed bases for linear value function approximation, and is competitive with learned proto-value functions.

Downloads

Published

2011-08-04

How to Cite

Konidaris, G., Osentoski, S., & Thomas, P. (2011). Value Function Approximation in Reinforcement Learning Using the Fourier Basis. Proceedings of the AAAI Conference on Artificial Intelligence, 25(1), 380-385. Retrieved from https://ojs.aaai.org/index.php/AAAI/article/view/7903

Issue

Section

AAAI Technical Track: Machine Learning