SM-SPSA

Release 0.0.1

Date Aug 29, 2024

SM-SPSA (stochastic matrix simultaneous perturbation stochastic approximation) is an extension of the SPSA algorithm to optimize a nonlinear objective function over the stationary distribution of a Markov chain. For more details on the SM-SPSA algorithm, please see the paper by Dieleman, Berkhout and Heidergott.

Citing

If you would like to cite this work, please see the README.md/CITATION.cff file on the GitHub repository for a suggested citation.

License

The GNU General Public License v3 (GPL-3) license is used. For more information, please see the GitHub repository.

Installation

User guide

API reference

Releases