Speech Processing by Computer

 

LECTURE 6a

NEURAL NETWORKS

 

 

Objectives

 

By the end of the session you should:

be able to justify an interest in neural networks

be able to contrast a real with an artificial neurone

be able to describe in general terms the computational processing that goes on inside an artificial neurone

appreciate that networks of artificial neurones can store, process and learn information

have gained experience with perceptrons learning some simple tasks

be able to list some application areas for neural networks in speech

 

Outline

 

1.      What is a neural network?

q       Computation inside an artificial neurone

q       Units, inputs, weights, activation, output

2.      Why are neural networks interesting

q       The appeal of connectionist processing

3.      Common architectures for networks

q       Hopfield network

q       Interactive Activation and Competition network

q       Feed-forward (perceptron) network

q       Dynamic (recurrent) network

q       Kohonen (self-organising) map

4.      Example of network learning

q       Perceptron learning

q       Self-organising map learning

 

Reading

 

J.L. McClelland, D.E.Rumelhart, G.E. Hinton, "The Appeal of Parallel Distributed Processing", in Parallel Distributed Processing Volume 1, ed Rumelhart & McClelland, MIT Press, 1986.