Skip to content

Instantly share code, notes, and snippets.

@drvinceknight
Last active September 24, 2016 10:20
Show Gist options
  • Select an option

  • Save drvinceknight/1ce4038b5bbc26ed5205313d21533837 to your computer and use it in GitHub Desktop.

Select an option

Save drvinceknight/1ce4038b5bbc26ed5205313d21533837 to your computer and use it in GitHub Desktop.

Paper accepted: "An Open Framework for the Reproducible Study of the Iterated Prisoner’s Dilemma "

Vince Knight

On the 31st of August, myself and 21 co-authors had an article accepted for publication in the journal of open research software. This list of authors includes mathematicians and software developers from various parts of the world as well as some of our own students.

Having 22 authors is not common in Mathematics but with the other core developers of the software in question, we decided that anyone who made a contribution (no matter how large or small) to the package would be invited to be a named author.

The software is a package for the study of the iterated prisoner's dilemma. This game theoretic area of research aims to understand emergence of cooperative behaviour in complex dynamics. It was originally started by Robert Axelrod who in 1980 invited submissions of computer code to play the prisoner's dilemma. Sadly, like a lot of research this code was not developed in a sustainable way and so a large amount of research on the prisoner's dilemma is not reproducible.

The software developed is a Python package, and is open source. All contributions are made using github (a popular repository for open source code) and this has not only lead to many strategies from the literature being reproduced but also novel strategies being developed. In particular the current two "best" strategies in the library are cleverly written strategies that are trained using particular swarm and genetic algorithms.

If you would like to read more about the library you can:

There have at present been more than 30 contributors to the library and this has been a wonderful experience for myself in open scientific research. I'm grateful to all contributors but to the core team (Owen Campbell, Marc Harper and Karol Langner) in particular. They have taught me a lot in terms of software development.

The hope is that this piece of software will contribute to the reproducibility of game theoretic research.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment