UTCS Artificial Intelligence
courses
talks/events
demos
people
projects
publications
software/data
labs
areas
admin
Discovering Multimodal Behavior in Ms. Pac-Man through Evolution of Modular Neural Networks (2016)
Jacob Schrum
and
Risto Miikkulainen
Ms. Pac-Man is a challenging video game in which multiple modes of behavior are required: Ms. Pac-Man must escape ghosts when they are threats and catch them when they are edible, in addition to eating all pills in each level. Past approaches to learning behavior in Ms. Pac-Man have treated the game as a single task to be learned using monolithic policy representations. In contrast, this paper uses a framework called Modular Multiobjective NEAT (MM-NEAT) to evolve modular neural networks. Each module defines a separate behavior. The modules are used at different times according to a policy that can be humandesigned (i.e. Multitask) or discovered automatically by evolution. The appropriate number of modules can be fixed or discovered using a genetic operator called Module Mutation. Several versions of Module Mutation are evaluated in this paper. Both fixed modular networks and Module Mutation networks outperform monolithic networks and Multitask networks. Interestingly, the best networks dedicate modules to critical behaviors (such as escaping when surrounded after luring ghosts near a power pill) that do not follow the customary division of the game into chasing edible and escaping threat ghosts. The results demonstrate that MM-NEAT can discover interesting and effective behavior for agents in challenging games.
View:
PDF
Citation:
IEEE Transactions on Computational Intelligence and AI in Games
, Vol. 8, 1 (2016), pp. 67--81.
Bibtex:
@article{schrum:tciaig16, title={Discovering Multimodal Behavior in Ms. Pac-Man through Evolution of Modular Neural Networks}, author={Jacob Schrum and Risto Miikkulainen}, volume={8}, journal={IEEE Transactions on Computational Intelligence and AI in Games}, number={1}, month={March}, pages={67--81}, url="http://www.cs.utexas.edu/users/ai-lab?schrum:tciaig16", year={2016} }
People
Risto Miikkulainen
Faculty
risto [at] cs utexas edu
Jacob Schrum
Ph.D. Alumni
schrum2 [at] southwestern edu
Areas of Interest
Evolutionary Computation
Game Playing
Multiobjective Optimization
Neuroevolution
Software/Data
MM-NEAT
Download at GitHub
Modular Multiobjective NEAT is a software fra...
2014
Demos
Multimodal Behavior in One Life Ms. Pac-Man
Jacob Schrum
2014
Labs
Neural Networks