Sunith Bandaru
Associate Professor of Production Engineering
School of Engineering Science
Choose a course instance to see course syllabus and admission requirements.
Application is done after nomination.
Computational Intelligence (CI) deals with methods and algorithms intended to solve NP-hard problems. Many problems in the real-world are ill-posed and therefore cannot be efficiently solved using deterministic methods. CI methodologies can be broadly classified into neuro, fuzzy and evolutionary techniques. These techniques emulate natural and biological systems to find approximate solutions in a reasonable time. CI also has a significant overlap with the fields of artificial intelligence and data-mining, where the focus is on discovering patterns and knowledge from data through machine learning.
This course will firstly introduce methodologies which fit the traditional definition of CI, namely: Artificial Neural Networks, Fuzzy Computation and Evolutionary Computation.
We will also learn to apply selected topics in machine learning, namely: supervised learning which includes Classification and regression trees, k-Nearest Neighbors, Neural Networks and Random Forests and Unsupervised learning which includes k-Means and Hierarchical Clustering.
Desirable Prerequisites are basic knowledge of vector and matrix operations and programming knowledge.