Value Function Approximation in Reinforcement Learning using the Fourier Basis

Authors

George Konidaris, University of Massachusetts Amherst
Sarah Osentoski, University of Massachusetts Amherst

Publication Date

2008

Abstract

We describe the Fourier Basis, a linear value function approximation scheme based on the Fourier Series. We empirically evaluate its properties, and 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 even though no extra experience or computation is required.

Comments

This paper was harvested from CiteSeer

Recommended Citation

Konidaris, George and Osentoski, Sarah, "Value Function Approximation in Reinforcement Learning using the Fourier Basis" (2008). Computer Science Department Faculty Publication Series. 101.
Retrieved from https://scholarworks.umass.edu/cs_faculty_pubs/101