jma127 / pybrium

Strategy and equilibria toolkit

pybrium: Strategy and equilibria toolkit

pybrium is a Python library for calculations invoing strategies and equilibria.

Installation

Install with the following one-liner:

$ pip install git+https://github.com/jma127/pybrium.git

Quickstart

Calculate the maximum-entropy equilibrium and Nash averaged ratings for a three-way tournament:

import torch
from pybrium import nash_average

winrates = torch.tensor([
    [0.5, 0.3, 0.5],
    [0.7, 0.5, 0.5],
    [0.5, 0.5, 0.5],
], dtype=torch.float32)
logits = torch.log(winrates) - torch.log(1.0 - winrates)
nash_equilibrium, nash_rating = nash_average(logits, steps=(2 ** 14))

See examples.py for more usage examples.

Contributing

Bugfixes and contributions are very much appreciated! Feel free to submit a pull request at any time.

Citing

If you use this library in your research, please consider citing it as follows:

@misc{ma2019pybrium,
  author = {Jerry Ma},
  title = {pybrium: Strategy and equilibria toolkit},
  year = {2019},
  journal = {GitHub repository},
  howpublished = {\url{https://github.com/jma127/pybrium}}
}

License

This source code is licensed under the BSD 3-clause license found in the LICENSE file in the root directory of this source tree.