Iterated Prisoner's Dilemma (IPD) - EPQ Research

Research Tool used for my EPQ project on the Iterated Prisoner's Dilemma

Description

This repository contains the source code for my EPQ dissertation investigating strategy performance in the Iterated Prisoner’s Dilemma. It includes:

• Custom implementations of adaptive Tit-for-Tat variants as well as variants of other strategies

• Tournament simulations (baseline, noisy, and evolutionary)

• Visualization tools for results analysis

Author

Lukas Sanitra (Lukisani)

Key Features

Three novel strategies:

• AdaptiveTitForTat

• AdaptiveTitForTat10

• WindowedForgivenessTFT

Payoff customization: Supports standard/modified reward structures

Noise simulation: Tests robustness to stochastic errors (5–20%)

Evolutionary dynamics: Simulating natural selection over successive generations

Credits

Classic strategy implementations (TitForTat, Alternator, etc.) follow behavioral specifications from the Axelrod tournament literature and were independently coded for this project. For reference implementations, see the Axelrod-Python library.

License

This project is licensed under the MIT License - see LICENSE.md for details.