Statistical Machine Learning

Prof. Pier Palamara, University of Oxford, Hilary term 2020

Course offered to Part B students (SB2b) and MSc students (SM4)

New: Revisions (Trinity Term)


You can access past exam questions via OXAM, which is on WebLearn [solutions here]. Note that the syllabus of this course is different from previous editions (e.g. it used to be a Part C course). Past exam questions for revision include:
  • Part B 2017 Q1, Q2 (excluding c), Q3
  • Part C 2016 Q1 (a-b), Q2 (a-b)
  • Part C 2015 Q1, Q2 (a-c), Q3 (b)
  • MSc 2017 Q7, Q8 (excluding b-c)
  • MSc 2014 Q7
  • MSc 2012 Q6, Q7
  • MSc and Part B 2018
  • MSc and Part B 2019
Revision Class 1 Week 2, recorded by Brian Zhang.
Plan: 2015 Part C Q1 and Q2 (a-c), Q3 (b) and selected parts of 2017 MSc Q7.
Revision Class 2 Week 3, recorded by Juba Nait Saada.
Plan: 2018 Part B exam..
Consultation Session 1 Week 4, Wednesday 20th May, 4pm-5pm with Pier Palamara.
To make the consultation sessions more efficient, please send your questions by email beforehand.
 (see email in General Info).
Consultation Session 2 Week 5, Wednesday 27th May, 10am-11am with Tom Rainforth.
To make the consultation sessions more efficient, please send your questions by email beforehand.
 (see email in General Info).

General Information


Course Team Pier Palamara (13 lectures + tutor), Tom Rainforth (3 lectures + tutor) Brian Zhang (tutor), Juba Nait Saada (tutor), David Rindt (TA), Tom Hadfield (TA), Zhongyi Hu (TA)
Email Lecturers:    
Tutors:    
TAs:    
Lectures HT Weeks 1-8, Statistics room LG.01, Tue 5pm-6pm, Thu 4pm-5pm. Note: Second lecture moved to Wed 9am-10am.
Classes HT weeks 3/4, 5/6, 7/8, TT Week 1 (details below)
Practicals MSc only: HT Week 4, Friday 11am-1pm (unassessed) and HT Week 8, Friday 11am-1pm (assessed, groups of 4)

Tutorials

Please contact to change group or any other matter related to class allocation.

Group Time (location) Class Tutor / TA
SB2B group 1 Monday 11.15am-12.45pm, weeks 3, 5, 7 in LG.03, TT1 in LG.05. Pier Palamara / Zhongyi Hu
SB2B group 2 Tuesday 11am-12.30pm, weeks 3, 5, 7 in LG.03, TT1 in LG.05 Brian Zhang / Tom Hadfield
SB2B group 3 Tuesday 2.30pm-4pm, weeks 3, 5, 7 in LG.03, TT1 in LG.05 Brian Zhang / Tom Hadfield
SB2B group 4 Wednesday 9.30am-11am, weeks 3, 5, 7, TT1 in LG.05 Juba Nait Saada / Zhongyi Hu
SB2B group 5 Monday 10am-11.30am, weeks 4, 6, 8 in LG.04, TT1 in LG.02 Tom Rainforth / David Rindt
SB2B group 6 Monday 11.30am-1pm, weeks 4, 6, 8 in LG.04, TT1 in LG.02 Tom Rainforth / David Rindt
SM4 (single group) Monday 3pm-4pm, weeks 3, 5, 7, TT1 in LG.01 Pier Palamara

Syllabus


Recommended prerequisites:
Part A A9 Statistics and A8 Probability. SB2a Foundations of Statistical Inference useful by not essential.

Aims and Objectives:
Machine learning studies methods that can automatically detect patterns in data, and then use these patterns to predict future data or other outcomes of interest. It is widely used across many scientific and engineering disciplines. This course covers statistical fundamentals of machine learning, with a focus on supervised learning and empirical risk minimisation. Both generative and discriminative learning frameworks are discussed and a variety of widely used classification algorithms are overviewed.

Synopsis
Visualisation and dimensionality reduction: principal components analysis, biplots and singular value decomposition. Multidimensional scaling. K-means clustering. Introduction to supervised learning. Evaluating learning methods with training/test sets. Bias/variance trade-off, generalisation and overfitting. Cross-validation. Regularisation. Performance measures, ROC curves. K-nearest neighbours as an example classifier. Linear models for classification. Discriminant analysis. Logistic regression. Generative vs Discriminative learning. Naive Bayes models. Decision trees, bagging, random forests, boosting. Neural networks and deep learning.

Reading
C. Bishop, Pattern Recognition and Machine Learning, Springer, 2007.
T. Hastie, R. Tibshirani, J Friedman, Elements of Statistical Learning, Springer, 2009.
K. Murphy, Machine Learning: a Probabilistic Perspective, MIT Press, 2012.

Further Reading
B. D. Ripley, Pattern Recognition and Neural Networks, Cambridge University Press, 1996.
G. James, D. Witten, T. Hastie, R. Tibshirani, An Introduction to Statistical Learning, Springer, 2013.


Slides


Slides will be made available as the course progresses and periodically updated. Please check back for updates.
You can use the material from last year (here) to get a sense of future lectures. The schedule below may change.
Please email us to report any typos or errors.

1 (PP) Introduction, unsupervised learning: exploratory data analysis Slides »
2 (PP) Principal component analysis Slides »
3 (PP) SVD, Biplots, multidimensional scaling, Isomap Slides »
4 (PP) Clustering (K-means, hierarchical) Slides »
5 (PP) Supervised learning: empirical risk minimization, regression (linear, polynomial), overfitting Slides »
6 (PP) Generalization, cross-validation, bias-variance trade-off Slides »
7 (PP) Regularization, (demo, source, data) gradient descent, classification: the Bayes classifier Slides »
8 (PP) Classification: linear and quadratic discriminant analysis (LDA, FDA) Slides »
9 (PP) QDA, logistic regression Slides »
10 (PP) Logistic regression (cont.), generative vs. discriminative learning, naïve Bayes Slides »
11 (PP) Naïve Bayes (cont.), K-nearest neighbors, evaluating performance, ROC curves Slides »
12 (PP) Neural networks Slides »
13 (PP) Boosting Slides »
14 (TR) Decision Trees Slides »
15 (TR) Bagging, random forests Slides »
16 (TR) Deep learning Slides »

Problem Sheets


For undergraduate students: hand in solutions in the pigeon holes labeled with your class time.
Class allocation details are on Minerva (accessible from Oxford network).

1 due at noon, January 31st Sheet » Solutions »
2 due at noon, February 14th Sheet » Solutions »
3 due at noon, February 28st Sheet » Solutions »
4 due at noon, April 24th Sheet » Solutions »


Resources for Python



Resources for R



Background Review Aids