A method for optimising a pre-existing mesh of tetrahedral ®nite elements is described. It is based on a series of mesh connectivity and node position searches of the landscape de®ning mesh quality. A Riemannian metric, re¯ecting the a posteriori error measure, is used to calculate element size and shape. A functional is de®ned which embodies both shape and size quality of an element with respe...

Irrelevant features and weakly relevant features may reduce the comprehensibility and accuracy of concepts induced by supervised learning algorithms. We formulate the search for a feature subset as an abstract search problem with probabilistic estimates. Searching a space using an evaluation function that is a random variable requires trading oo accuracy of estimates for increased state explora...

Much human learning appears to be gradual and unconscious, suggesting a very limited form of search through the space of hypotheses. We propose hill climbing as a framework for such learning and consider a number of systems that learn in this manner. We focus on CLASSIT, a model of concept formation that incrementally acquires a conceptual hierarchy, and MAGGIE, a model of skill improvement tha...

Classification Association Rule Mining (CARM) systems operate by applying an Association Rule Mining (ARM) method to obtain classification rules from a training set of previously-classified data. The rules thus generated will be influenced by the choice of ARM parameters employed by the algorithm (typically support and confidence threshold values). In this paper we examine the effect that this ...

This paper presents new results showing that a very simple stochastic hill climbing algorithm is as good or better than more complex metaheuristic methods for solving an oversubscribed scheduling problem: scheduling communication contacts on the Air Force Satellite Control Network (AFSCN). The empirical results also suggest that the best neighborhood construction choices produce a search that i...

The recent explosion of high dimensionality in datasets for several domains has posed a serious challenge to existing Bayesian network structure learning algorithms. Local search methods represent a solution in such spaces but suffer with small datasets. MMHC (MaxMin Hill-Climbing) is one of these local search algorithms where a first phase aims at identifying a possible skeleton by using some ...

The tool switching problem (ToSP) is well known in the domain of flexible manufacturing systems. Given a reconfigurable machine, the ToSP amounts to scheduling a collection of jobs on this machine (each of them requiring a different set of tools to be completed), as well as the tools to be loaded/unloaded at each step to process these jobs, such that the total number of tool switches is minimiz...

The objective of this thesis is to investigate how to improve Particle Swarm Optimization by hybridization of stochastic search heuristics and by a Self-Organized Criticality extension. The thesis will describe two hybrid models extending Particle Swarm Optimization with two aspects from Evolutionary Algorithms, recombination via breeding and gene flow restriction via subpopulations. A further ...

Object-oriented systems that undergo repeated addition of functionality commonly suffer a loss of quality in their underlying design. This problem must often be remedied in a costly refactoring phase before further maintenance programming can take place. Recently search-based approaches to automating the task of software refactoring, based on the concept of treating object-oriented design as a ...