This library implements a computational model of Elinor Ostrom's Institutional Analysis and Development framework. It includes the interpreter to the Action Situation Language (ASL) and the game engine to automatically generate extensive-form games from ASL descriptions.

Requirements

ngames requires a working installation of the following:

• Python 3
• SWI-Prolog
• The PySwip package

Usage

For the time being, the ngames package requires to download a local copy of the source code. The path to the package should then be appended to your Python path (see how).

See the examples folder for detailed illustrations of the framework in action. To analyse your own scenarios, you should create your ASL description in three distinct files:

• agents.pl
• states.pl
• rules.pl

Then, to construct the extensive-form game semantics of your description, it is enough to call:

>>> build_full_game(<path_to_ASL_description>, <identifier>, (threshold=..., max_rounds=...))

See the documentation for further details.

References

Ostrom, E. (2005). Understanding Institutional Diversity. Princeton University Press.

Montes, N., Osman, N., & Sierra, C. (2021). Enabling Game-Theoretical Analysis of Social Rules. In Artificial Intelligence Research and Development (Vol. 339, pp. 90–99). IOS Press. https://doi.org/10.3233/FAIA210120

Montes, N., Osman, N., & Sierra, C. (2022). A Computational Model of Ostrom’s Institutional Analysis and Development Framework. Artificial Intelligence (Vol. 311). Elsevier. https://doi.org/10.1016/j.artint.2022.103756