NPCDS/MITACS 2006 "Learning" lectures and labs

• Day 1:
  • Lecture 1: Introduction - Russ Steele (pdf)
  • Lecture 2: Bootstrapping/CV/Bayesian methods for penalizing complexity - Russ Steele (pdf)
  • Lecture 3: Software introduction - Hugh Chipman
    • Basic facts about R: What is R? on the R webpage
    • A sample R session giving some basic commands without too much discussion.
  • Lab: Introduction to R: Day 1 lab. - Hugh Chipman, Russ Steele, others
  • Additional supplementary materials for R:
    • R reference card - a 4-page card summarizing some of R's functionality.
    • Preliminaries for R - instructions on how to install R, not necessary for use during the workshop.
    • Introduction to R, a detailed (99 page) description of R's functionality.
    • Hugh's Data Analysis using R, a 3-hour introductory session for students with little or no background in R. I'm not sure I'll use any of this.
• Day 2:
  • Lecture 1 - From biological to artifical neural networks -- Helmut Kroger (ppt | pdf) Note that pdf is in black and white as colour slides didn't convert well.
  • Lecture 2 - More on Neural Networks: Algorithmic Issues -- Doina Precup (pdf)
  • Lecture 3.0 - Supervised learning with Artificial Neural Nets, part I -- Antonio Ciampi (pdf)
  • Lecture 3.5 - Hugh Chipman - Neural Nets: A simple example (pdf)
  • Lecture 4 - Supervised learning with Artificial Neural Nets, part II -- Antonio Ciampi (pdf)
  • Lab: Neural networks - Hugh Chipman, others
• Day 3: Model-based clustering
  • Lecture 1 - Russ Steele (pdf)
  • Lecture 2 - Russ Steele (pdf)
  • Lab 1 - Russ Steele (sml.summerschool.mbc.R)
  • Lab 2 - Russ Steele (MBC.raftery.R)
• Day 4:
  • Lecture 1 & 2: Background and Statistical Learning Interpretation - Ji Zhu (pdf)
  • Lab: Support vector machines - Ji Zhu
• Day 5: Manifold learning
  • Lecture -- Yoshua Bengio : (pdf)