rssalessio / PrivacyStochasticSystems

Code used in "Minimizing Information Leakage of Abrupt Changes in Stochastic Systems" , by Alessio Russo, [e-mail] .

PrivacyStochasticSystems Library

Code used in "Minimizing Information Leakage of Abrupt Changes in Stochastic Systems" - Alessio Russo, Alexandre Proutiere, 2021.

Author: Alessio Russo (PhD Student at KTH - [e-mail])

License

Our code is released under the MITlicense (refer to the LICENSE file for details).

Requirements

To use the library you need atleast Python 3.7. To run the notebooks you need to install jupyter notebooks.

Other required dependencies:

• NumPy
• SciPy
• CVXPY
• DCCP
• Matplotlib
• TQDM

If you have Conda installed you may install the required packages using the command conda env create --file=env.yaml

Usage/Examples

Check the notebook files. If possible, use the MOSEK solver. Alternatively, one can use ECOS (which is included in cvxpy). That may require some fine-tuning.

• Linear system - full information
• Linear system - limited information
• MDP with 3 states

Citations