Prospects of the Interdisciplinary Course “Computational Intelligence” in Engineering Education

Abstract

AbstractThe experience of teaching lecture courses on genetic algorithms, evolutionary computing, and computational intelligence at a number of universities in Ukraine and St. Petersburg at the undergraduate and graduate levels is presented. The content of the new course “Computational Intelligence” (CI) is presented, which covers 5 main paradigms: artificial neural networks, evolutionary computing, fuzzy systems, swarm algorithms, and artificial immune systems. All 5 paradigms are successfully applied in solving many scientific and technical problems separately. However, modern trends require the development of hybrid systems where combinations of these paradigms are used. Hybrid systems are able to take advantage of the individual strengths of specific paradigms and compensate for the shortcomings of various CI paradigms. Various forms of hybridization of the components of computational intelligence are considered. On this basis, new powerful algorithms are developed to solve complex problems. The interdisciplinary character of CI is noted, which determines the prospects for its application in the development of cyber-physical systems using the processes of generalization, abstraction, and association inspired by nature. CI can be an important tool for multidisciplinary education and research.KeywordsComputational intelligenceNeural networksEvolutionary computingSwarm algorithmsArtificial immune systemsHybrid systemsInterdisciplinary educationCyber-physical systems