Sei sulla pagina 1di 1

Introduction to Neural networks

Objectives: to present the basic network architectures for classification and regression; to introduce the neural networks as means for computational learning; to give design methodologies for artificial neural networks; to provide knowledge for network tunning and overfitting avoidance; to demonstrate neural network applications on real-world tasks.

Outcomes: design single and multi-layer feed-forward neural networks; explain the differences between networks for supervised and unsupervised learning; explain the behavior of radial-basis function networks; understand unsupervised learning using Kohonen networks; understand training of recurrent Hopfield networks; perform algorithmic training of various neural networks; analyse the performance of neural networks. develop multilayer neural networks for regression; implement training algorithms for multilayer neural networks; achieve overfitting avoidance in nonlinear regression.construct nonlinear models for regression; analyse the results from nonlinear data modeling.

Prerequests:
EE351K (Probability and Stats) or equivalent. Knowledge of Matlab will be helpful. M 340L is helpful but not needed.

Potrebbero piacerti anche