neat-nagi-python

The code in this repository is an implementation of the framework "Neuroevolution of Artificial General Intelligence" and is part of Kristoffer Olsen Master's Thesis project at the University of Oslo. It is part of the SOCRATES project, and any further updates and development on the implementation will be done in this current fork of the original repository.

Installation

git clone https://github.com/SocratesNFR/neat-nagi-python.git
export PYTHONPATH="/path/to/neat-nagi-python":$PYTHONPATH

Evolving and simulating an adaptive spiking neural network

You start in the 'scripts' folder. Run NEAT algorithm. After finished, you extract the genome from the generate pickle file inside 'data' folder. The extracted genome can be simulated with 'simulation_*.py'.

cd scripts
python run_neat_*.py
python extract_genome.py
python simulation_*.py

Evolved adaptive spiking neural networks

If you want to simulate already evolved spiking neural networks. They are located in 'evolved_nets' folder.

Publications

Preprint with conceptual framework

Sidney Pontes-Filho and Stefano Nichele. "A Conceptual Bio-Inspired Framework for the Evolution of Artificial General Intelligence." arXiv preprint arXiv:1903.10410 (2019).

Kristoffer Olsen's Master thesis

Kristoffer Olsen. "Neuroevolution of Artificial General Intelligence." Master thesis, University of Oslo (2020).

Preprint with developed framework and first results

Sidney Pontes-Filho, Kristoffer Olsen, Anis Yazidi, Michael A. Riegler, Pål Halvorsen and Stefano Nichele. "Towards the Neuroevolution of Low-level Artificial General Intelligence." arXiv preprint arXiv:2207.13583 (2022).

Citing this work

If you use NAGI for academic research, please cite the following paper:

@article{pontes2022nagi,
  title={Towards the Neuroevolution of Low-level Artificial General Intelligence},
  author={Pontes-Filho, Sidney and Olsen, Kristoffer and Yazidi, Anis and Riegler, Michael and Halvorsen, Pål and Nichele, Stefano},
  journal={arXiv preprint arXiv:2207.13583},
  year={2022}
}