Journal of Computer Science & Technology

About This Issue

This particular section of the Journal of Computer Science and Technology is devoted to Intelligent Systems. As David Fogel [1] said "Intelligence may be defined as the capability of a system to adapt its behavior to meet desired goals in a range of environments". Intelligent behavior then requires prediction, for adaptation to future circumstances requires predicting those circumstances and taking appropriate action.

From the beginning of Artificial Intelligence researchers attempted to insert intelligence into systems, replicating human behavior. The first approaches were known as expert systems where the participation of a human expert was required to create and update a knowledge data base containing a set of inference rules, used later on for decision making.

At the time of the renaissance of neural networks in the 80’s, human behavior was again attempted at the neural connection level and neural nets, trained under supervised or unsupervised learning techniques, can today learn how to take decisions in complex environments or efficiently classify big amounts of data.

Rather than seek to generate system intelligence by replicating humans, either in the rules they may follow or in their neural connections, an alternative approach to generate intelligent systems is to simulate evolution. Evolutionary computation has recently been recognized as a research field. Under this name main approaches to simulation of various aspects of evolution are known: genetic algorithms, evolution strategies, evolutionary programming and genetic programming. Unnecessary boundaries between these main approaches are presently disappearing because the essential substance of evolution (reproduction, random variation, competition and selection of individuals) is found in their fundamental commonality and successful techniques from one approach are adopted on other approaches.

Some of the current theoretical trends involve research on adaptation in the search or parameter space, coevolution, multiple recombination methods, specialized operators, and hybridization. Some current applications involve multiobjective optimization, robotics, biomodeling, cancer diagnosis, engineering design, scheduling, stock market, signal processing, parallel and distributed processing, network planning, forecasting and manufacturing optimization.

The strength of evolutionary computation is supported by the experience and quality of results obtained every day when facing complex problems. This fact justify its current progress. Even though we are far to arrive to a complete theoretical foundation of this emerging field, many researchers contribute regularly to shorten distance to this goal.

Some of the attempts, theoretical and practical, to build intelligent systems are published in the current issue of the Journal of Computer Science and Technology.

Enjoy them !!

[1] Fogel D. B., Evolutionary Computation: Towards a new Philosophy of Machine Intelligence. Piscataway NJ: IEEE. 1995

Raúl Gallard (rgallard@unsl.edu.ar)

Universidad Nacional de San Luis, Departamento de Informática.

[Home]