Skip to content
Go to file

Latest commit


Git stats


Failed to load latest commit information.


To learn how to compile/use, skip straight to

Copyright (c) 2014 The University of Texas at Austin and 2016 Southwestern University. All rights reserved. Refer to LICENSE.txt for detailed license information. Also see copyright.txt for copyright information about the included Ms. Pac-Man code.


MM-NEAT stands for Multiobjective Modular Neuro-Evolution of Augmenting Topologies. It is inspired by the original NEAT, but also incorporates multiobjective evolution via NSGA-II, and supports several forms of modular neural networks. Support for the fitness shaping technique Targeting Unachieved Goals (TUG) is also included. Indirect encoding via HyperNEAT is also supported, as is interactive evolution of several types of interesting artifacts (pictures, sounds, animations) using CPPNs. The code was originally developed by Jacob Schrum ( while at the University of Texas at Austin, but has since been improved upon by several undergraduate students at Southwestern University in Georgetown, TX, where Dr. Schrum is currently a professor in the department of Math and Computer Science. Links to publications and demos further explaining the code are available at Dr. Schrum's personal webpage:

A list of publications is also in

More information on NEAT is available in:

K. O. Stanley and R. Miikkulainen, "Evolving Neural Networks Through Augmenting Topologies." Evolutionary Computation, 10(2):99-127, 2002. URL:

More information on HyperNEAT is available in:

K. O. Stanley, D. B. D'Ambrosio, and J. Gauci, "A Hypercube-Based Encoding for Evolving Large-Scale Neural Networks." Artificial Life, 15(2):185–212, 2009. URL:

Information on NSGA-II is available in:

K. Deb, S. Agrawal, A. Pratap, and T. Meyarivan, "A Fast Elitist Non-dominated Sorting Genetic Algorithm for Multi-Objective Optimization: NSGA-II". Parallel Problem Solving from Nature (PPSN VI), pp. 849-858, 2000. URL:

A precursor to MM-NEAT is the BREVE Monsters software package, available at:

MM-NEAT was developed primarily to evolve multimodal behavior in Ms. Pac-Man, and therefore includes (modified) code for the Ms. Pac-Man simulator created for the Ms. Pac-Man vs. Ghosts Competitions. The original version of this code does not seem to be available anymore, but a newer version associated with the latest competition can be downloaded at:

For further instructions on how to run this code, see

For information on the different types of interactive evolution tasks in the code, see



Modular Multiobjective (Hyper) Neuro-Evolution of Augmenting Topologies: Java code for evolving intelligent agents in Ms. Pac-Man, Tetris, and more!




No packages published


You can’t perform that action at this time.