Computational Intelligence Topics


The following topics will be covered in this module.

Topic 1: Optimization

The application domain for this module is optimization. Therefore, the module starts with an overview of optimization theory and concepts. Formal definitions of optimization concepts are provided, and different classes of optimization problems are defined and discussed.

Topic 2: Computational Swarm Intelligence: Particle Swarm Optimization

Particle swarm optimization (PSO) is one of the most efficient and most used computational swarm intelligence algorithms. This topic will explore the PSO in detail, starting with the basic PSO, convergence behavior of PSO, roaming behavior of PSO, a number of single-solution PSO algorithms, heterogeneous PSO algorithms and self-adaptive PSO approaches. This is followed by studies of PSO algorithms for large-scale optimization, discrete-valued optimization problems, constrained optimization, multi- and many-objective optimization, multi-modal optimization, and dynamic optimization problems.

Topic 3: Computation Swarm Intelligence: Ant Algorithms

The very first Computational Intelligence algorithms were inspired by ant foraging behavior. This topic focus on these algorithms, referred to as ant colony meta-heuristics. In addition ant algorithms inspired by the cemetery organization behavior and division-of-labor behavior of ants are also studied.

Topic 4: Evolutionary Computation

This topic focuses on different classes of evolutionary algorithms, i.e. algorithms inspired by evolution in nature. Genetic algorithms, genetic programs, evolutionary programming, evolutionary strategies, differential evolution, cultural evolution, and co-evolution algorithms are studied. In addition, estimation of distribution algorithms will be discussed.

Topic 5: Hyper-Heuristics

Hyper-heuristics are optimization algorithms which solves optimization problems, and while doing so, also searches for the best algorithms from a pool of algorithms to find optimal solutions. The focus is on selection hyper-heuristics.