Evolutionary dynamics of finite populations in games with polymorphic fitness equilibria

Sevan G. Ficici, Jordan B. Pollack
2007 Journal of Theoretical Biology  
RESEARCH Multi-agent systems, computational game theory, machine learning, stochastic search, evolutionary computation, adaptive behavior, robotics, artificial life, complex systems, dynamical systems, computer music. University of Rochester Eastman School of Music, Rochester NY M.A. Music Theory (1990) • Master's Thesis: Computer Analysis of Surface Detail in Tonal Music Cleveland Institute of Music, Cleveland OH B.M. Music Theory (1988) Honor Roll 1984-1988 • Bachelor's Thesis: Compositional
more » ... echniques in Haydn's Opus 20 and Opus 33 String Quartets RESEARCH EXPERIENCE Harvard University School of Engineering and Applied Sciences, Cambridge MA Post-Doctoral Research Fellow in Computer Science with Avi Pfeffer, Barbara Grosz, Stuart Shieber, 2005-Present • Conducted extensive human-subjects trials for research in modeling human strategic reasoning • Designed probabilistic computational models of human strategic reasoning and used them to build computer agents capable of successful interaction with people in strategic multi-agent settings • Researching new, compact representations of complex strategic situations involving many agents • Researching multi-agent learning of cooperative behavior by selfish autonomous agents • Helping design and build Colored Trails, a DARPA/NSF funded multi-agent system testbed • Interdisciplinary research involving behavioral economics Affinnova, Cambridge MA Senior Research Scientist, 2001-2004 (part-time) Research Consultant, 2000-2001 Market research and product design and packaging for consumer packaged goods • Reported to Chief Technology Officer • Responsible for design of innovative, state-of-the-art interactive evolutionary algorithm technology (see Patents) This technology has been used to design products and packaging for internationally recognized consumer brands • Greatly scaled ability of interactive evolutionary algorithms to extract information from human feedback • Consulted with colleagues in operations to assist them with system configuration • Met with prospective clients and partners to help promote Affinnova technology •
doi:10.1016/j.jtbi.2007.03.004 pmid:17466341 fatcat:eby26pyvmbfjzag6qkgat6ulue