Song-Ju Kim: “Natural Intelligence: Novel method for harnessing the computational power of physical phenomena”
12:15 〜 13:00
CiNet 1F Conference Room
National Institute for Materials Science (NIMS)
Host PI : Ferdinand Peper
Our society comprises a collection of individuals, each of whom handles decision-making tasks that are abstracted as computational problems of finding the most profitable option from a set of options that stochastically provide rewards. Society is expected to maximize the totalrewards, while the individuals compete for common rewards. Such collective decision making is formulatedas the “competitive multi-armed bandit problem (CBP),” requiring a huge computationalcost that rapidly grows as a function of the numbers of individuals and options.
Herein, we demonstrate a prototype of an analog computer that efficientlysolves CBPs by exploiting the physical dynamics of numerous fluids in coupled cylinders. This device enables the maximization of the total rewards for the society without paying the conventionally required computational cost; this is because the fluids estimate the reward probabilities of the options for the exploitation of past knowledge and generate random fluctuations for the exploration of new knowledge. Our results suggest that to optimize the social rewards, the utilization of fluid-derived natural fluctuations is more advantageous than applying artificial external fluctuations. Our analog computing scheme is expected to trigger further studies for harnessing the huge computational power of natural phenomena for resolving a wide variety of complex problems in modern information society.
The Friday Lunch Seminar is CiNet's main regular meeting series, held every week at 12:15 in the beautiful main lecture theatre on the ground floor at CiNet. The talks are typically 40mins long and orientated towards an inter-disciplinary audience. They are informal, social, and most people bring their own lunch to eat during the talk. They are open to anyone who is feeling curious and wants to come, regardless of where you work.