Overview
COM4509/COM6509 Machine Learning and Adaptive Intelligence 201415
Exam 2014/15
You can find solutions to this year’s paper here.
Exam
Information on the Exam is now available.
Course Overview
This unit aims to provide a deep understanding of the fundamental technologies underlying modern artificial intelligence. In particular it will provide foundational understanding of probability and statistical modelling, supervised learning for classification and regression, and unsupervised learning for data exploration. The teaching consists of two hours of lectures and one of lab classes each week. The lectures are on Tuesdays, the labs on Fridays. The teaching schedule and venue for each week are given below:
 Tuesday 910 MAPPLT12; Tuesday 1113 ADBA04 Lab Class
 Tuesday 910 MAPPLT12; Tuesday 1113 ADBA04 Lab Class
 Tuesday 910 MAPPLT12; Tuesday 1113 AT1012 Lab Class
 Tuesday 910 MAPPLT12; Tuesday 1113 ADBA04 Lab Class
 Reading Week
 Tuesday 910 MAPPLT12; Tuesday 1113 ADBA04 Lab Class
 Tuesday 910 MAPPLT12; Tuesday 1113 MAPPF110 Lab Class
 Tuesday 910 MAPPLT12; Tuesday 1113 MAPPF110 Lab Class
 Tuesday 910 MAPPLT12; Tuesday 1113 MAPPF110 Lab Class
 Tuesday 910 MAPPLT12; Tuesday 1113 MAPPF110 Lab Class
 Reading Week
 Tuesday 910 MAPPLT12; Tuesday 1113 MAPPF110 Lab Class
Prerequisites
For this course we assume that you have seen linear algebra before as well as probability and calculus. The course is in Python, but we will not be ‘teaching’ the language. The assumption is that you have enough prior experience of programming to pick up the language as you go. If you are unfamiliar with these concepts you might want to refresh yourself with this lab class on basic machine learning in python with the notebook.
Recommended text book
The main course recommended text is Rogers and Girolami’s “A First Course in Machine Learning”. Also useful is Bishop, Pattern Recognition and Machine Learning. Most lectures will provide references to these text, and it will help a lot if you read the relevant sections in your own time.
A further publicly available text is Hastie et. al, The Elements of Statistical Learning.
Prerequisites
You are expected to have familiarity with basic probability and linear algebra. We will use Python and the ipython notebook on the course, so you are expected to be comfortable with adapting to a new programming environment without specific tuition.
Lecture Slides: Schedule
The material for the lectures will be posted below before each lecture (including audio and screen capture, where possible). We aim to put up the materials for each week’s lectures at the beginning of week.

Week 1: Uncertainty and Probability Introduction to Pandas, Jupyter and Probability

Week 2: Objective Functions and Recommender Systems Recommender systems Lab

Week 3: Linear Algebra and Regression Linear Algebra and Regression in Python

Week 4: Basis Functions Basis Functions Lab

Week 5: Reading Week

Week 6: Generalization Generalization Lab

Week 7: Bayesian Regression Bayesian Regression Lab

Week 8: Dimensionality Reduction Dimensionality Reduction Lab

Week 9: Classification: Naive Bayes Classification Lab

Week 10: Classification: Logistic Regression Classification Lab

Week 11: Reading Week and Question and Answer

Week 12: Gaussian Processes Gaussian Process Lab