0%

COMP3009 - 0.Module Overview

1. Teaching

1.1. Theoretical Part:

  • 2 x 2 hours/week online videos
  • lecture activities

1.2. Practical Part: Lab and 3 group Coursework

  • ANN - Artifical Neural Networks
  • Decision Trees
  • Support Vector Machines

2. Assessment

  • 2h Exam(70%)
    • Contents from all lectures examinable, as well as the relevant chapters in the two core text (Bishop and Duda & Hart)
    • Guest lectures can be used as examples but not examinable in detail.
    • Three questions - no optional question
  • One coursework (30%)
    • Reports on 3 Lab topics
    • Peer-review evaluation

3. Resources

  • Core text:

    • Machine Learning & Pattern Recognition (Christopher Bishop)
  • Additional Reading:

    • Machine Learning (Tom Mitchell)
    • Pattern Classification (Duda & Hart)

4. Structure

  1. Introduction
  2. Linear Regression
  3. Mathematics for Machine Learning
  4. Linear Classification
  5. Decision Trees
  6. Artificial Neural Networks I
  7. Artificial Neural Networks II
  8. Deep Learning and Convolutional Neural Networks
  9. Evaluating Hypotheses
  10. Unsupervised Learning
  11. Data mining
  12. Dimensionality Reduction
  13. Kernel Methods
  14. Support Vector Machines
  15. Bayesian Learning
  16. Probabilistic Graphical Models
  17. Sequential Data
  18. Advanced Topics
  19. Exam Revision