Tad Gonsalves AI Lab
Prof. Tad Gonsalves, PhD
Department of Information & Communication Sciences,
Faculty of Science & Technology,
Sophia University,
Tokyo, Japan
RESEARCH
ARTIFICIAL INTELLIGENCE APPLICATIONS
Artificial Intelligence (AI) is a well-researched branch of computer science. After a long period of low-lying, it has suddenly sprung back to life and is expected to make a breakthrough in designing algorithms that will be very close to human thinking, intuition and creativity. For the time being, my lab is not too concerned with the theoretical development of AI, but rather with its diverse applications.
The research in my lab ideals with the application of Artificial Intelligence (AI) techniques and algorithms to various disciplines. In particular, we focus on the following three areas:
DESIGN OF EXPERT SYSTEMS
With my under-grad and graduate students, I am trying to develop knowledge-based systems, which are traditionally called Expert Systems. These systems contain a large knowledge-based (similar to a database) acquired from domain experts. They have an inference capability to refer to the knowledge and give advice to users in the domain of interest. The current research area is how to embed learning mechanisms and semantics in these systems.
NATURE INSPIRED ALGORITHMS
Nature-inspired algorithms are robust algorithms that work with ill-defined problems and produce near-optimal results in reasonable amount of time without making an exhaustive search of the search space.
Our research is concentrated on two major areas involving the nature-inspired Algorithms
1. Evolutionary Algorithms
These algorithms are based on the Darwinian metaphor. Living beings continuously evolve to adjust themselves to the ever-changing environmental conditions. The fully evolved living beings possess various optimized facets which make them successful in adapting to their environment. Evolutionary Algorithms mimic the Darwinian paradigm of the survival of the fittest. They begin by randomly generating a population of feasible solutions to a given optimization problem. The population is then subjected to a handful of operators like selection, crossover, and mutation to give rise to fitter (optimized) solutions through a large number of generation cycles.
2. Swarm Intelligence
A swarm is a large number of homogenous, unsophisticated agents that interact locally among themselves and their environment without any central control or management. The collective behavior of self-organized, but decentralized natural or artificial systems that leads to the solution of complex problems is called Swarm Intelligence. The individuals that make up the swarm are often extremely simple agents, that lack memory, intelligence or even awareness of one another. By following simple rules like sticking together and avoiding collision, they give rise to a form of emergent intelligence.
MACHINE LEARNING
Machine Learning is a rapid developing area of AI. In recent years, many awe-inspiring Machine Learning Applications have been developed by academia as well as industry. Here are just a few applications we have developed in our lab.