Spike-based decision learning of nash equilibria in two-player games

Humans and animals face decision tasks in an uncertain multi-agent environment where an agent's strategy may change in time due to the co-adaptation of others strategies. The neuronal substrate and the computational algorithms underlying such adaptive decision making, however, is largely unknown. We propose a population coding model of spiking neurons with a policy gradient procedure that successfully acquires optimal strategies for classical game-theoretical tasks. The suggested population reinforcement learning reproduces data from human behavioral experiments for the blackjack and the inspector game. It performs optimally according to a pure (deterministic) and mixed (stochastic) Nash equilibrium, respectively. In contrast, temporal-difference(TD)-learning, covariance-learning, and basic reinforcement learning fail to perform optimally for the stochastic strategy. Spike-based population reinforcement learning, shown to follow the stochastic reward gradient, is therefore a viable candidate to explain automated decision learning of a Nash equilibrium in two-player games.

Culture, risk factors and mortality: can Switzerland add missing pieces to the European puzzle?

BACKGROUND: The aim was to compare cause-specific mortality, self-rated health (SRH) and risk factors in the French and German part of Switzerland and to discuss to what extent variations between these regions reflect differences between France and Germany. METHODS: Data were used from the general population of German and French Switzerland with 2.8 million individuals aged 45-74 years, contributing 176 782 deaths between 1990 and 2000. Adjusted mortality risks were calculated from the Swiss National Cohort, a longitudinal census-based record linkage study. Results were contrasted with cross-sectional analyses of SRH and risk factors (Swiss Health Survey 1992/3) and with cross-sectional national and international mortality rates for 1980, 1990 and 2000. RESULTS: Despite similar all-cause mortality, there were substantial differences in cause-specific mortality between Swiss regions. Deaths from circulatory disease were more common in German Switzerland, while causes related to alcohol consumption were more prevalent in French Switzerland. Many but not all of the mortality differences between the two regions could be explained by variations in risk factors. Similar patterns were found between Germany and France. CONCLUSION: Characteristic mortality and behavioural differentials between the German- and the French-speaking parts of Switzerland could also be found between Germany and France. However, some of the international variations in mortality were not in line with the Swiss regional comparison nor with differences in risk factors. These could relate to peculiarities in assignment of cause of death. With its cultural diversity, Switzerland offers the opportunity to examine cultural determinants of mortality without bias due to different statistical systems or national health policies.