Class schedule and syllabus - EEE511 - Fall 2007

Last revised 12-02-2007  (Press shift-reload in your browser to force loading the latest version.)

This is a tentative schedule. Changes are very likely.

 

August 20

  • Introduction to artificial neural networks (Ch 1)
  • Artificial neural networks in relation to other disciplines
  • Application areas of ANNs
  • Motivating demonstrations
  • Assignment 1 for next class: Get the primers from the web page and read them!
  • Assignment 2 for next class: Email the questionnaire to the instructor!

August 22

  • More motivating demonstrations
  • Neuron models
  • Overview of principles and methods of neural computing (Ch. 2)

August 27

  • Overview continued (Ch. 2)
  • Single layer networks (Ch. 3)
  • Overview of some unconstrained optimization methods (Ch. 3)

August 29

  • Overview of some unconstrained optimization methods (Ch. 3)
  • Single layer networks: LMS (Ch. 3)
  • 1st homework out

September 5

  • Principles of pattern recognition
  • Single layer networks: Perceptrons  (Ch. 3)

September 10

  • Multilayer perceptrons and back-propagation (Ch. 4)

September 12

  • Representational capabilities of MLPs, training issues (Ch. 4)
  • 1st homework due (Sept 14)
  • 2nd homework out (Sept 16)

September 17

  • Generalization, avoiding overfitting with multilayer networks (Ch. 4)
  • Regularization, Cross-Validation (Ch. 4)

September 19

  • Homework 1 solutions review

September 24

  • Optimization methods for MLPs (Ch. 4)
  • Radial basis function networks, Introduction (Ch. 5)

September 26

  • Radial basis function networks, Exact interpolation  (Ch. 5)
  • Radial basis function networks, Regularization (Ch. 5)
  • Radial basis function networks, Model selection  (Ch. 5)

October 1

  • (reserved)

October 3

  • (reserved)
  • 2nd homework due (Oct 5)

October 8

  • Homework 2 solutions review
  • 3rd homework out

October 10

  • Radial basis function networks, Learning basis functions (Ch. 5)
  • Start seriously thinking about the topic of the final project

October 15

  • The self-organizing map, Introduction  (Ch. 9)
  • The self-organizing map, the basic algorithm  (Ch. 9)

October 17

  • The self-organizing map, properties and analysis of the algorithm (Ch. 9)
  • The self-organizing map: Tree-Structurd SOMs, non-static network topologies.

October 22

  • The self-organizing map: Sequential data representation.
  • The self-organizing map: Application cases.

October 24

  • The self-organizing map: More application cases.
  • Learning Vector Quantization (Ch 9)

October 29

  • Support vector machines: Maximizing the margin, VC-dimension (Ch. 6)
  • Support vector machines: Separable and nonseparable cases, Quadratic optimization (Ch. 6)
  • Support vector machines: The kernel trick (Ch. 6)
  • 3rd homework due

October 31

  • Homework 3 solutions review
  • Deadline for the final project proposal is today.
  • 4th homework out

November 5

  • Support vector machines: The kernel trick (Ch. 6)
  • Support vector machines: Regression, examples (Ch. 6)

November 7

  • Temporal processing with feedforward networks (Ch 13)
  • Recurrent networks (Ch 15)

November 12

  • No class --- Veteran’s Day observed

November 14

  • Recurrent networks continued (Ch 15)
  • Information-theoretic methods with neural networks (Ch 10)
  • 4th homework due (Nov 17, 11am)
  • 5th homework out (Nov 14)

November 19

  • Homework 4 solutions review (briefly)
  • Information-theoretic methods with neural networks (cont), including Independent Components Analysis  (Ch 10)
  • What did we not cover?
  • Recapitulation of the main concepts
  • Reassembling the big picture

November 21

November 26

November 28

December 3