CSC321 Spring 2009 - Tutorials

• January 8
  Demonstrations of some neural network learning algorithms
  The first two demonstrations are simply for your enjoyment. They will not be on any tests so do not read the papers unless you want to. The third demonstration will be explained at great length towards the end of the course. You can try reading about it now, but you may find it very hard to understand before you have the necessary background.
  Demo 1: Glovetalk: A net that converts hand-movements to speech. pdf
  Demo 2: NeuroAnimator: A net that learns to emulate a physical system. pdf
  Demo 3: A net that learns to recognize really bad handwritten characters flash demo
  

• January 15
  Introduction to Matlab for complete novices.
  Download the first matlab tutorial: first Matlab tutorial as .pdf.
  

• January 22
  Using matlab for learning (very helpful for assignment 1)
  The tutorial will show how to use Matlab to implement some of the simple learning algorithms in the lectures. It is intended mainly for Matlab novices, but it will also help you understand the code in assignment 1.
  Download the second Matlab tutorial: second Matlab tutorial as .pdf.
  Matlab code used in second tutorial:
  classdata1.m
  classdata2.m
  graddescent_simple.m
  graddescent_vectorized.m
  perceptron.m
  regressiondata.m
  showclassdecision.m
  

• Jan 29
  Tutorial on: Essential Concepts from Probability Theory.
  .pdf file for probability tutorial
  

• February 5
  Post mortem on assignment 1
  Explanation of assignment 2

• February 12
  Review of all material in the first 11 lectures.
  This is your last chance to ask questions before the midterm.

• February 26
  Tutorial Lecture 1 on Support Vector Machines
  The material in this lecture will not be in the exam or assignments
  (notes as .ppt ) (notes for all browsers)) (notes as .ps, 4 per page))
  

• March 5
  Post mortem on assignment 2
  Explanation of assignment 3

• March 12
  Tutorial Lecture 2 on Support Vector Machines
  The material in this lecture will not be in the exam or assignments
  (notes as .ppt ) (notes for all browsers)) (notes as .ps, 4 per page))
  

• March 19
  Post mortem on assignment 3
  Explanation of assignment 4
  Some worked examples of the computations used in Hidden Markov Models
  

• March 26
  Extra lecture on ISOMAP & Stochastic Neighbor Embedding (not on exam)
  (notes as .ppt ) (notes for all browsers)) (notes as .ps, 4 per page))
  

• April 2
  Post mortem on assignment 4
  Explanation of assignment 5

• April 9
  Review of all material in lectures 13-25.

CSC321 - Computation In Neural Networks: || www.cs.toronto.edu/~hinton/csc321/