robupy is an open-source Python package for finding worst-case probabilities in
the context of robust decision making. It aims to collect algorithms, which find for
different construction methods for the ambiguity set, the worst-case distribution as
fast as possible.
The first algorithm implemented, applies to an ambiguity set constructed with the Kullback-Leibler divergence function. It reduces the selection to a one-dimensional minimization problem. This algorithm was developed and described in:
Nilim, A., & El Ghaoui, L. (2005). Robust control of Markov decision processes with uncertain transition matrices. Operations Research, 53(5): 780–798.
You can install robupy via conda with
$ conda config --add channels conda-forge
$ conda install -c opensourceeconomics robupyPlease visit our online documentation for tutorials and other information.
If you use robupy for your research, do not forget to cite it with
@Unpublished{robupy.2020,
author = {{robupy}},
title = {A {P}ython package for robust optimization},
year = {2020},
url = {https://github.com/OpenSourceEconomics/robupy/releases/1.1.1},
}