Balancing selection pressures, multiple objectives, and neural modularity to coevolve cooperative agent behavior

Title

Balancing selection pressures, multiple objectives, and neural modularity to coevolve cooperative agent behavior

Creator

Schrum, Jacob

Rollins, Alex C.

Date

2018-12-10

Date Available

2018-12-10

Date Issued

2017-07

Identifier

Alex C. Rollins and Jacob Schrum. 2017. Balancing selection pressures, multiple objectives, and neural modularity to coevolve cooperative agent behavior. In Proceedings of the Genetic and Evolutionary Computation Conference Companion (GECCO '17). ACM, New York, NY, USA, 75-76. DOI: https://doi.org/10.1145/3067695.3075979

uri

https://collections.southwestern.edu/s/suscholar/item/243

Abstract

Previous research using evolutionary computation in Multi-Agent Systems indicates that assigning fitness based on team vs. individual behavior has a strong impact on the ability of evolved teams of artificial agents to exhibit teamwork in challenging tasks. Such research made use of single-objective evolution, but when multiobjective evolution is used, populations can be subject to individual-level objectives, team-level objectives, or combinations of the two. This paper explores the performance of cooperatively coevolved teams of agents controlled by artificial neural networks subject to these types of objectives. Because of the tension between individual and team behaviors, multiple modes of behavior can be useful, so the effect of modular neural networks is also explored. Results demonstrate that fitness rewarding individual behavior is superior to fitness rewarding team behavior, despite being applied to a cooperative task. However, networks with multiple modules can discover intelligent behavior, regardless of which type of objectives are used.

Language

English

Publisher

GECCO '17 Proceedings of the Genetic and Evolutionary Computation Conference Companion

Subject

Computing methodologies

Machine learning

Machine learning approaches

Neural networks

Bio-inspired approaches

Generative and developmental approaches

Type

Article