About Me

My photo
Tokyo, Japan
The Hon. Dr. Hideyasu M. R. "Hide" Sasaki is a computer scientist working in Gov't of Japan for Big Data Initiative and Catholic lawyer admitted to practice in New York, the United States.

Thursday, May 11, 2017


I've got a letter of acceptance of my abstract for an invited talk at FSDM 2017 in Taiwan, that is the 3rd International Conference on Fuzzy Systems and Data Mining at National Dong Hwa University (NDHU) in Taiwan on November 24-27, 2017. My talk is entitled as "Collective Intelligence Modeling in Time Sensitive Decision Making". Its abstract goes like the following:


Collective intelligence is a nature-inspired problem solving technique that has been highly appreciated in machine learning and data mining of Big Data. Time-sensitive decision making highlights the next challenge of Big Data learning by deep neural networks. In this talk, after short briefing where collective intelligence survives at the advent of deep learning modeling, I will present two collective intelligence models for time-sensitive decision making. The first model is on bilateral decision making, and is formulated by introducing collective intelligence about human-machine interaction that dramatically accelerates the decision-making speed. Moreover, that model is enhanced into multilateral decision making under time constraint.


The second model applies collective intelligence, which is found in modeling human-machine interaction under time constraint, to very big biological data. In analytics of behaviors of the biological Big Data, we discovered how tiny creatures like ants make a wise choice in moving between an old and a new nest within a very short period of time. The ant’s moving is modeled and reduced into a simple probabilistic distribution that shows a very powerful way collective intelligence works in time-sensitive decision making. Through discussing the two collective intelligence models, we would offer exemplary cases on where collective intelligence, furthermore heuristics, that machine learning and data mining have discovered until the present time, contributes to the future research of time-sensitive deep learning.

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.