Hugo Terashima-Marín

Abstract: This paper describes an investigation of solving Examination Timetabling Problems (ETTPs) with Genetic Algorithms (GAs) using a non-direct chromosome representation based on evolving the configuration of Constraint Satisfaction methods. There are two aims. The first is to circumvent the problems posed by a direct chromosome representation for the ETTP that consists of an array of events in which each value represents the timeslot which the corresponding event is assigned to. The second is to show that the adaptation of particular features in both the instance of the problem to be solved and the strategies used to solve it provides encouraging results for real ETTPs. There is much scope for investigating such approaches further, not only for the ETTP, but also for other related scheduling problems.

TL;DR: In this article, the authors present an approach for dynamic split strategies in Constraint Problem Solving using Fuzzy Extension of Description Logic (DELL) for solving the Stereo Correspondence Problem.

TL;DR: A GA-based method that produces general hyper-heuristics that solve two-dimensional regular (rectangular) and irregular (convex polygonal) bin-packing problems and a variable-length representation is presented.

TL;DR: A Neurobiologically Motivated Model for Self-organized Learning and Using Boolean Differences for Discovering Ill-Defined Attributes in Propositional Machine Learning.

TL;DR: A hyper-heuristic methodology is described that can generate a fast, deterministic algorithm capable of producing results comparable to that of using the best problem-specific heuristic, and sometimes even better, but without the cost of trying all the heuristics.