PyNEAT: A Python implementation of NEAT

July, 2020

PyNEAT is a python implementation of the NeuroEvolution of Augmenting Topologies (NEAT) algorithm created by Kenneth O’Stanley. PyNEAT began as an original implementation of NEAT, and then was modified to use the same interface as NEAT-Python to reuse various reporting and configuration features they had developed. However, the core behaviour of the NEAT algorithm implemented in PyNEAT is significantly different to that of NEAT-Python, and is much closer to the original specification of NEAT.



  1. IVCNZ
    Evaluating Learned State Representations for Atari
    Adam Tupper, and Kourosh Neshatian
    In 2020 35th International Conference on Image and Vision Computing New Zealand (IVCNZ), Nov 2020
  2. GECCO
    Evolving Neural Network Agents to Play Atari Games with Compact State Representations
    Adam Tupper, and Kourosh Neshatian
    In Proceedings of the 2020 Genetic and Evolutionary Computation Conference Companion, Nov 2020
  3. Evolutionary Reinforcement Learning for Vision-Based General Video Game Playing
    Adam Tupper
    University of Canterbury, Aug 2020