Value Function Approximation in Reinforcement Learning Using the Fourier Basis

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.

Cite

Text

Konidaris et al. "Value Function Approximation in Reinforcement Learning Using the Fourier Basis." AAAI Conference on Artificial Intelligence, 2011. doi:10.1609/AAAI.V25I1.7903

Markdown

[Konidaris et al. "Value Function Approximation in Reinforcement Learning Using the Fourier Basis." AAAI Conference on Artificial Intelligence, 2011.](https://mlanthology.org/aaai/2011/konidaris2011aaai-value/) doi:10.1609/AAAI.V25I1.7903

BibTeX

@inproceedings{konidaris2011aaai-value,
  title     = {{Value Function Approximation in Reinforcement Learning Using the Fourier Basis}},
  author    = {Konidaris, George Dimitri and Osentoski, Sarah and Thomas, Philip S.},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2011},
  pages     = {380-385},
  doi       = {10.1609/AAAI.V25I1.7903},
  url       = {https://mlanthology.org/aaai/2011/konidaris2011aaai-value/}
}