Sistemas Inteligentes
Objectives
The course will present artificial intelligence techniques for extracting usefull knowledge from data. More specifically, the course will introduce in details concepts such as Optimization and machine Learning and it will focus on stochastic heuristic methods like, among the others, Genetic Algorithms, Particle Swarm Optimization and Neural Networks.
General characterization
Code
100097
Credits
6.0
Responsible teacher
Leonardo Vanneschi
Hours
Weekly - Available soon
Total - Available soon
Teaching language
Portuguese. If there are Erasmus students, classes will be taught in English
Prerequisites
No requirement
Bibliography
Machine Learning. Tom Mitchell; Genetic programming: on the programming of computers by means of natural selection. J. Koza; 0; 0; 0
Teaching method
Classes are partitioned into theoetical classes and practical classes.
Theoretical classes will be held using the board and slides.
In the practical classes students will implement the algorithms presented in the theoretical classes.
Evaluation method
First epoch: project (30%) and oral exam (70%)
Subject matter
Optimization problems: definition of problem and instance of a problem
Search space, neighborhood structure and related concept
No free lunch theorem
Local search
Simulated annealing
Genetic algorithms
Genetic Programming
Semantic genetic programming
Pareto dominance
Multi objective optimization (NSGA II)
Machine Learning
Classification and clustering
Performance of a Classifier
Generalization and overfitting
Feature selection
Artificial Neural Networks
Programs
Programs where the course is taught: