Introductory Applied Machine Learning
Course Homepage
Lecturers: Victor Lavrenko and
Charles Sutton
TA:
Sean Moran
Class rep:
Evgeniya (Jenny) Sotirova
Office Hours: TBA
The goal of this course is to introduce students to basic algorithms
for learning from examples, focusing on classification and clustering
problems.
This is a level 9 course intended for MSc students and 3rd year
undergraduates.
For an overview of the planned course topics, see the
course descriptor and
the introductory handout.
Lectures
2pm Mondays (AT LT1) and Thursdays (AT LT3)
Resources
Discussion Forum
A
discussion
forum has been made for this course. This is monitored by both
lectures and the TA, so if you ask questions here, you are likely to
get a much faster response than if you email the lecturers
individually. (However, if you have issues that should be kept
confidential, then of course please do email the course lecturers.)
Assignments
There will be one assignment worth 25% of the mark for the course,
the question sheet a111.pdf
is available.
The assignment will be due on Tue 22 Nov 2011, 4pm.
You are required to submit both an electronic copy and a
manual copy to the ITO by the deadline. The deadline is
strictly enforced.
The data files are available:
train_faces.arff,
train_faces_clean.arff,
train_faces_clean_best.arff,
train_faces.mat,
train_faces_clean.mat,
val_faces.arff,
val_faces_best.arff
val_faces.mat,
test_faces.arff,
test_faces_best.arff
You will also need the following Matlab visualization code:
draw_eigenfaces.p,
draw_faces.p,
draw_pixels.p,
reconstruct_face.p,
sortem.p,
pc_evectors.p,
You must save these all of these files to a directory on your local machine, pointing the Matlab path to that directory (see the assignment for
instructions).
Important: In some parts of the assignment you are required to use Matlab for visualization. As there are a limited number of Matlab licenses
in the University, please try to keep the duration of your Matlab session to a minimum. Ensure that you close Matlab once you are finished so that
another student can use the license.
Tutorials
Tutorials will be 4 in weeks 3, 5, 7 and 9.
[groups].
- Mon 5pm - AT 4.14
- Tue 9am - AT 5.03
- Tue 3pm - AT 4.14
- Tue 5pm - AT 4.14
- Wed 9am - AT 4.14
- Wed 11am - AT 4.14
- Wed 2pm - AT 4.14
- Thu 10am - AT 4.14
- Thu 11am - AT 4.14
- Thu 3pm - AT 5.07
- Fri 11am - AT 5.07
- Fri 1pm - AT 5.07
- Fri 3pm - AT 3.03
Lab Classes
Labs will be in weeks 4,6,8 and 10.
[groups].
- Monday 1pm in AT 5.05
- Tuesday 5pm in AT 5.05
- Wednesday 5pm in AT 5.05
- Thursday 1pm in AT 5.05
- Friday 5pm in AT 5.05
Week-by-Week listing
(This list is subject to change.)
Week 1
Lectures:
Introduction
[slides],
mathematical preliminaries
[slides],
Thinking about Data
[slides]
Readings: Textbook chapters 1, 2, 7.1, 7.2
Tutorials No tutorials in week 1 (nor week 2)
Mathematical preliminaries These
Supplementary Mathematics notes
are from the old Learning from Data course. They are
more difficult than what we will need for IAML, but if
you are happy with them you should have no problem with
the IAML maths level.
Week 2
Lectures:
Basic Probability and Estimation
[slides]
Naive Bayes Classification
[slides]
Web Resources
Geomaths notebook on discrete and continuous probability distributions,
Geomaths notebook on matrices
Tutorials: No tutorials in week 2
Week 3
Lectures:
Decision trees
[slides]
Tutorials: Basic probability and data representation [Solutions]
Week 4
Lectures:
Nearest neighbor method,
KD tree
[slides]
Linear regression
[slides]
[1up]
Lab 1: Naive Bayes classification
Week 5
Lectures:
Overfitting and generalization
[slides]
[1up],
Linear classification and logistic regression (Updated 3 Nov)
[slides]
[1up],
Tutorials: Naive Bayes and Decision Trees [Solutions]
Week 6
Lectures:
Optimization
[slides]
[1up],
Perceptrons (updated 28 Oct)
[slides]
[1up],
Support vector machines
[slides]
[1up],
Readings: SVM handout
Lab 2: Decision trees, linear regression
Week 7
Lectures:
Support vector machines, Part II
[slides]
[1up],
K-means clustering
[slides],
Tutorials: Logistic regression
[Solutions]
Week 8
Lectures:
Mixture models and the EM algorithm
[slides],
Dimensionality reduction
[slides],
Additional demo: Eigenfaces (not examinable)
Lab 3: Logistic regression, SVM, Feature Selection
Week 9
Lectures:
Evaluation
[slides],
Tutorials: SVMs, Clustering
[Solutions]
Week 10
Lectures:
Hierarchical clustering
[slides],
Neural networks
[slides]
(Updated 5 Dec)
Lab 4: PCA, Clustering, Evaluation
This page is maintained by
Victor Lavrenko
and Charles Sutton.
