4Interactive Model Learning

In this chapter, we present a study based on an evolutionary framework to explore what would be a reasonable compromise between interaction and automated optimization in finding possible solutions for a complex problem, namely the learning of Bayesian network (BN) structures, a non-deterministic (NP)-hard problem1 between where user knowledge can be crucial to distinguish polynomial solutions of equal fitness but very different physical meaning. Even though several classes of complex problems can be effectively tackled with evolutionary computation, most possess qualities that are difficult to directly encode in the fitness function or in the individual’s genotype description. Expert knowledge can sometimes be used to integrate the missing information, but new challenges arise when searching for the best way to access it: full human interaction can lead to user-fatigue, while a completely automated evolutionary process can miss important contributions by the expert. For our study, we developed a graphical user interface (GUI)-based prototype application that lets an expert user guide the evolution of a network by alternating between fully interactive and completely automatic steps. Preliminary user tests were able to show that despite still requiring some improvements with regard to its efficiency, the proposed approach achieves its goal of delivering satisfying results for an expert user in food model case studies. The work described in this chapter ...

Get Evolutionary Algorithms for Food Science and Technology now with the O’Reilly learning platform.

O’Reilly members experience live online training, plus books, videos, and digital content from nearly 200 publishers.