CZ3205  Neural Computing

Last updated 19 March 2001

References:  (1) "Neural Networks", 2nd ed, S Haykin;  (2) "Neural Computation and Self-Organizing Maps", H Ritter et al;  (3) "Introduction to the Theory of Neural Computation", J Hertz et al. 

A/P Wang Jian-Sheng's part (Chapters 6-9) - Lectures are on 14 Feb, 7, 14, 21 March, 2001, 9:00-12:00 for the 1st two lectures, and 10:00-12:00 for the rest three in S17 07-41.

Chapter 1: Introduction

    A multi-disciplinary point of view 
Application of neural networks 
Historical notes

Chapter 2: Properties and Models of Neurons

    Overview of the brain 
Properties of single neurons 
    Synaptic integration and neuron models

Chapter 3: Linear Neural Network

    Mach Bands and Lateral Inhibition 
    Demonstrating Lateral Inhibition
    Linear Associative Network

Chapter 4: The Perceptron 

    Perceptron and classification
    Perceptron Learning algorithm and convergence theorem
Limitations of Perceptrons
    Widrow-Hoff learning

Chapter 5: Back-Propagation Algorithms

    Multilayer networks
    Back-propagation algorithms

Midterm Mini-Review

Chapter 6: Unsupervised Learning 

    Principal components analysis
    Kohonen's self-organizing maps

Chapter 7: Energy and Neural Network 

    Hopfield's networks
    Memory capacity

Chapter 8: Statistical Physics and Stochastic Learning Machines

    Simulated Annealing
    Gibbs Sampling
    Boltzmann Machine

Chapter 9: Neurodynamics

    Nonlinear autoassociative neural network
    Brain-State-in-Box model
    Associative computation

Chapter 10: Selected Topics