Cognitive: Foundation of the basic computational intelligence concepts in the areas of neural networks, statistical learning, fuzzy logic and evolutionary computation. Comprehension of the ideas behind self-organization, competitive learning, reinforcement learning. Conception of the fuzzy logic, fuzzy sets and fuzzy systems principles. Foundation of the evolutionary concepts in optimization. Exposition to the plethora of computational intelligence applications in multimedia information analysis and retrieval, intelligent control, robotics and neurofuzzy systems.
Skills: Promoting analytic and implementation skills in the taught computational intelligence fields (fuzzy theory, evolutionary computation, swarm intelligence and neural networks) and their application to real problems. Combining computational intelligence techniques for complex real problems like computer vision, multimedia information retrieval, robotic control.
Course Content (Syllabus)
Neural Networks overview. Associative memories. Hopfiled recurrent neural networks. Self-organizing neural networks. Kohonen maps and competitive learning. Simulated annealing and Boltzmann machine. Deep Learning. Deep Convolutional Neural Networks. Deep Autoencoders. Deep Recurrent Networks. Reinforcement learning neural networks. Fuzzy logic, fuzzy learning nad fuzzy systems. Evolutionary and genetic algorithms. Swarm intelligence algorithms. Genetic programming, Socio-cognitive and mimetic algorithms. Applications in pattern recognition and robotics. Neuro-fuzzy systems. Hybrid computational intelligence systems in signal, image and video analysis.
Computational Intelligence, Deep Learning, Fuzzy Learning, Evolutionary Algorithms
Course Bibliography (Eudoxus)
1. ΥΠΟΛΟΓΙΣΤΙΚΗ ΝΟΗΜΟΣΥΝΗ, Σ. ΤΖΑΦΕΣΤΑΣ, Σ.ΤΖΑΦΕΣΤΑΣ, 2η Εκδοση, ΑΘΗΝΑ, 2008
2. ΝΕΥΡΩΝΙΚΑ ΔΙΚΤΥΑ ΚΑΙ ΜΗΧΑΝΙΚΗ ΜΑΘΗΣΗ, 3η ΕΚΔΟΣΗ, S. Haykin, Παπασωτηρίου, 2010, Αθήνα.