Interactive evolutionary computation: fusion of the capabilities of EC optimization and human evaluation

Abstract

We overview the research on interactive evolutionary computation (IEC). The IEC is an EC that optimizes systems based on human subjective evaluation. First, the definition and features of the IEC are described. Then, the overview of the IEC research follows. The overview consists of two parts: application and interface researches. The IEC application fields in this survey include graphic art and animation, 3-D CG lighting, music, editorial design, industrial design, face image generation, speech processing and synthesis, hearing aid fitting, virtual reality, media database retrieval, data mining, image processing, control and robotics, food industry, geophysics, education, entertainment, and so on. The interface research to reduce human fatigue in this survey includes interfaces of fitness input and display based on fitness prediction, accelerating EC convergence especially in early EC generations, combinations of interactive and normal ECs, and active user intervention. Finally, we discuss the IEC from the point of the future research direction of computational intelligence. The biggest feature of this paper is to wholly survey them with 200 IEC papers rather than to carefully select representative papers to show the status quo of the IEC researches.

Adaptive user interfaces in complex supervisory tasks

“…the operator as much as 1.01 KHz can be. Takagi [68] proposed to use NN (Neural Network ) in order to model f and this can correspond to user model in Benyon [7].…”

“…Our approach to the problem is finding the mapping between two spaces of interest Takagi [68]. These are psychological and parameter spaces.…”

Steps Toward a Modular Library for Turning Any Evolutionary Domain into an Online Interactive Platform

“…Together, the Picbreeder and IESoR domains aim to cover a broad range of interests inside the ALife community. Picbreeder is a proxy for domains that are more suited towards Interactive Evolutionary Computation (IEC) (Takagi, 2001) and difficult to optimize. In contrast, IESoR is a control-based domain that runs multiobjective search to discover new ambulatory creatures, i.e.…”

Combining Structural Analysis and Multi-Objective Criteria for Evolutionary Architectural Design

“…Human interaction has allowed EC to be applied to problems such as music and computer graphics, and to act as an exploratory tool as opposed to its primary function as an optimiser. A more complete list of interactive applications can be found in [13] and [14].…”