Grana M., Duro R., D'Anjou A., Wang P.P. (eds.) Information Processing with Evolutionary Algorithms. From Industrial Applications to Academic Speculations

Springer, 2005, -339 p.The last decade of the 20th century has witnessed a surge of interest in numerical, computation-intensive approaches to information processing. The lines that draw the boundaries among statistics, optimization, artificial intelligence and information processing are disappearing, and it is not uncommon to find well-founded and sophisticated mathematical approaches in application domains traditionally associated with ad-hoc programming. Heuristics has become a branch of optimization and statistics. Clustering is applied to analyze soft data and to provide fast indexing in the World Wide Web. Non-trivial matrix algebra is at the heart of the last advances in computer vision.
The breakthrough impulse was, apparently, due to the rise of the interest in artificial neural networks, after its rediscovery in the late 1980s. Disguised as ANN, numerical and statistical methods made an appearance in the information processing scene, and others followed. A key component in many intelligent computational processing is the search for an optimal value of some function. Sometimes, this function is not evident and it must be made explicit in order to formulate the problem as an optimization problem. The search often takes place in high-dimensional spaces that can be either discrete, or continuous or mixed. The shape of the high-dimensional surface that corresponds to the optimized function is usually very complex. Evolutionary algorithms are increasingly being applied to information processing applications that require any kind of optimization. They provide a systematic and intuitive framework to state the optimization problems, and an already well-established body of theory that endorses their good mathematical properties. Evolutionary algorithms have reached the status of problem-solving tools in the backpack of the engineer. However, there are still exciting new developments taking place in the academic community. The driving idea in the organization of this compilation is the emphasis in the contrast between already accepted engineering practice and ongoing explorations in the academic community.
After the seminal works of Holland, Goldberg and Schwefel, the field of evolutionary algorithms has experienced an explosion of both researchers and publications in both the application-oriented and the fundamental issues. It is obviously difficult to present in a single book a complete and detailed picture of the field. Therefore, the point of view of this compilation is more modest. Its aim has been to provide a glimpse of the large variety of problems tackled with evolutionary approximations and of the diversity of evolutionary algorithms themselves based on some of the papers presented at the Frontiers on Evolutionary Algorithms Conference within JCIS 2002 and complemented with some papers by well-known authors in the field on topics that were not fully covered in the sessions. Following the general trend in the field, most of the papers are application-oriented. However, we have made an effort to include some that refer to fundamental issues as well as some that provide a review of the state of the art in some subfield.
As the subtitle From industrial applications to academic speculations suggests, the organization of the compilation follows an axis of nearness to practical applications. We travel from industrial day-to-day problems and practice to the more speculative works. The starting collection of papers is devoted to immediate applications of clear economical value at present.Adaptive Business Intelligence Based on Evolution Strategies: Some Application Examples of Self-Adaptive Software
Extending the Boundaries of Design Optimization by Integrating Fast Optimization Techniques with Machine Code Based, Linear Genetic Programming
Evolutionary Optimization of Approximating Triangulations for Surface Reconstruction from Unstructured 3D Data
An Evolutionary Algorithm Based on Morphological Associative Memories for Endmember Selection in Hyperspectral Images
On a Gradient-based Evolution Strategy for Parametric Illumination Correction
A New Chromosome Codification for Scheduling Problems
Evolution-based Learning of Ontological Knowledge for a Large-Scale Multi-Agent Simulation
An Evolutionary Algorithms Approach to Phylogenetic Tree Construction
Robot Controller Evolution with Macroevolutionary Algorithms
Evolving Natural Language Grammars
Evaluating Protein Structure Prediction Models with Evolutionary Algorithms
Learning Decision Rules by Means of Hybrid-Encoded Evolutionary Algorithms
Evolvable Hardware Techniques for Gate-Level Synthesis of Combinational Circuits
The Evolutionary Learning Rule in System Identification
Current and Future Research Trends in Evolutionary Multiobjective Optimization
Genetic Algorithms with Limited Convergence
Evolution with Sampled Fitness Functions
Molecular Computing by Signaling Pathways
Strategy-Oriented Evolutionary Games: Toward a Grammatical Model of Games
Discrete Multiphase Particle Swarm Optimization