TUTORIAL
Hideyuki Takagi (Kyushu University, Japan)

He received the degrees of Bachelor, Master, Doctorate Degrees in 1979, 1981, and 1991. He worked for the Central Research Laboratories of Panasonic in 1981-1995, and is an Associate Professor of Kyushu Institute of Design since 1995 and now works for Kyushu University after two universities merged in 2003. He was a visiting researcher at UC Berkeley in 1991-1993.

He is interested in Computational Intelligence (CI), especially cooperation of several CI techniques and human. Currently, his interest focuses on Interactive Evolutionary Computation (IEC) that aims the cooperation of human and EC. He is one of the most active researchers in IEC research community as shown in his publication, organizing sessions at conferences, giving invited talks and lectures, receiving paper awards, and other academic activities. He received eight academic awards.

He is the Vice President of IEEE Systems, Man, and Cybernetics Society (SMCS) in 2006-2007 and 2008-2009, a registered lecturer of the SMCS Distinguish Lecturer Program in 2006-2007 and 2008-2009, the Chair of SMCS Technical Committee on Soft Computing, and an Associate Editor of IEEE Trans. on SMC Part B.

See his detail bio at here.

Tutorial Abstract:

Interactive Evolutionary Computation

Interactive Evolutionary Computation (IEC) is a method for optimizing target systems based on human knowledge, experiences, preference, intuition, and/or KANSE in general. There are many systems that it is hard or impossible to design their fitness functions for the optimization and therefore we cannot apply conventional optimization methods including conventional evolutionary computation (EC) framework. Such optimization tasks include designing cute robotics motions, better sound quality of hearing aids, image enhancement for medical diagnostics, and others.

Before turning into the introduction of IEC, we first explain the big picture of computation intelligence research in the 20th Century and how IEC works in this century from the view point of the research trend of computational intelligence.

Secondly, we overview IEC applications to understand how to use IEC in wide application areas. I roughly categorize the application areas into three: artistic applications, engineering applications, and others, and explain many IEC applications in these areas. Some of them are: artistic computer graphics design, editorial or industrial design, web design, melody and rhythm composition, face design, sound and image processing, speech synthesis, data mining, robotics and control, media database retrieval, MEMS design, geological simulation, game, and others.

Thirdly, we overview IEC applications of reducing IEC user fatigue. As IEC user must cooperate with a tireless computer, many iterative evaluations causes human fatigue. To minimize the IEC user fatigue, several trials have been conducted. Some of them are: improving IEC interface, accelerating EC convergence in early generations, making IEC user's evaluation model for simulating IEC process, and others.

Fourthly, we explain the recent trend of IEC research. One is to use IEC to analyze human mind indirectly by analyzing the target systems optimized based on IEC user's mental scale. The second is combination of IEC with evolutionary multi-objective optimization, The third one is expanding IEC framework.

This tutorial includes Q&A for not only understanding IEC but also advising how to apply IEC to the application tasks that audience are tackling now.