HYBRID CONNECTIONIST SYSTEMS IN RESEARCH AND TEACHING

Authors
Citation
Lc. Jain, HYBRID CONNECTIONIST SYSTEMS IN RESEARCH AND TEACHING, IEEE aerospace and electronic systems magazine, 10(3), 1995, pp. 14-18
Citations number
16
Categorie Soggetti
Engineering, Eletrical & Electronic","Aerospace Engineering & Tecnology
ISSN journal
08858985
Volume
10
Issue
3
Year of publication
1995
Pages
14 - 18
Database
ISI
SICI code
0885-8985(1995)10:3<14:HCSIRA>2.0.ZU;2-B
Abstract
There is a tremendous world-wide interest in the development of the th eory and application of artificial intelligence (AI) and related techn iques both in tertiary institutions and in industry. This paper presen ts a brief survey of overseas AI teaching programs together with a rev iew of the teaching program in the University of South Australia. In a ddition, post-graduate projects undertaken by the Knowledge-Engineerin g Systems Group (KES), and sponsored by the Defence, Science and Techn ology Organisation (DSTO), an discussed. These involve expert systems, neural networks, fuzzy logic and genetic algorithms. It is shown by u sing a classical exclusive-OR problem that it is useful to use hybrid techniques which offer fast convergence in optimization problems. Alth ough much of this work is still at an early stage, sufficient progress has been made to demonstrate the merit of using AI and related techni ques to help create solutions to these problems.