42035 - Neural and Evolutionary Computation (CNE) [UPC]

Type: S3 Course
Semester: Fall
ECTS: 6
Teaching Points: 15
Offer: Annual
Responsible Unit: CS-UPC
Responsible: René Alquézar
Language: English
Requirements:

GOALS

The main objective of this course is to provide the students with a wide understanding of the state of the art in neural and evolutionary computation techniques, including from the different optimization methods for training neural networks to some recently introduced models, such as support vector machines and heterogeneous neural networks.

CONTENTS

1. Neural computation

Introduction to optimization methods
Derivative based optimization methods
Incremental methods for neural network construction
ayesian neural networks
Recurrent neural networks
Heterogeneous neural networks
Support vector machines

2. Evolutionary computationç

Introduction to evolutionary computationç
Evolutionary optimization methods
Neural network training using evolutionary algorithms