Speaker: Prof. Cheol-Min Ghim Ulsan National Institute of Science and Technology
Time: 2014-05-29 14:00-2014-05-29 15:00
Venue: FIT 1-222

Abstract:

We discuss the persistence and proliferation of cooperation in social dilemma games in finite yet spatially structured populations. To elucidate the structural determinants of dynamics in an evolutionary game setting, we generate ensembles of networks with controlled graph characteristics, where individuals residing on each node play a general 2x2 game iteratively with their immediate neighbors. One prominent observation is that the intrinsic heterogeneity prevailing in social networks has negative impacts on the fixation of a single cooperating invader in a sea of defectors. Furthermore, as long as the cooperators (defectors) have eventually taken over the whole population, the fixation (loss) of the cooperator allelecomes significantly earlier in power-law graphs than in the exponential random graphs with the same mean degree. For graphs obtained by degree-preserving rewiring of an exponential random graph, the fixation probability is almost independent of the rewiring scheme. In the meantime, for power-law graphs, the fixation probability is strongly dependent on the degree-degree correlation, showing monotonic decrease of fixation probability with increasing assortativity. These findings are compared to the mean-field treatment of frequency-dependent replicator dynamics to shed light on the linkage between evolutionary game dynamics and directed percolation-type critical phenomena shown in magnetic spin systems.

Short Bio:

Cheol-Min Ghim is a theoretical physicist by training, holding Ph.D. in physics from Seoul National University. Since 2009 he has been professor in Biomedical Engineering and Physics at Ulsan National Institute of Science and Technology, working on biological information processing in molecular, cellular or population levels. In particular, he is interested in bridging the gap between the biochemical information processing and its in silico counterparts. From the biology side, he aims to rationalize the robustness of complex information processing networks in the light of structure-function duality. From the engineering side, he has been working on restoring the precision and accuracy of physical sciences in, ultimately, synthetically engineered cells.