General:

Syllabus

Things you will know by the end of this course

Project suggestions

Revised syllabus

Lecture notes:

"Supervised learning in single-stage feeforward networks"

"Supervised multilayer feeforward networks - backpropagation"

"Linear Hebbian learning and PCA"

"Attractor neural networks"

"A probability primer"

"Bayesian probability theory"

Problem sets:

Problem set # 1

Problem set #2

• matlab file: sparsedata.mat
• matlab file: eegdata.mat

Problem set #3

• Risset's endless glissando
  Chordgliss.aiff - play using playaiff on SGI's
  Chordgliss.au - for SUN and NEXT
  Chordgliss.wav - for Macs & PC's
  Chordgliss.mat - *use this version to load into Matlab* (shorter --> less memory)
• image: log_and_rocks.mat

Problem set #4

• reduced image: log_and_rocks256.mat
  reduced image: log_and_rocks128.mat

Problem set #5

• face images: faces.mat

Problem set #6

• intructions for running the simulation
• patterns.mat - patterns X O S
• hopnet.m - script for simulating network
• genpat.m - script for generating patterns (optional)