Module Syllabus and Resources 2016

Taken by students from:

Code	Degree Title	Year of study	core or option	credits
P-F3P4	Complexity Science MSc	1	option	12 CATS
P-F3P5	Complexity Science MSc+PhD	1	option	12 CATS
P-F3P6/7	Erasmus Mundus Masters in Complex Systems	1	option	6 ECTS

Context: This is the second opening module of the Complexity DTC taught programme.

Module Aims:

The problem of searching for regularities in data and using such regularities to better understand the properties of the system from which the data were obtained has a long history in science. In recent years, there has been a massive increase in our ability to collect data from complex systems of all kinds, from living cells to the Internet. The field of machine learning is concerned with computational methods for analyzing such data. This course introduces key concepts in machine learning, with a particular emphasis on approaches relevant to “reverse-engineering” complex systems with multiple interacting components.

link to Learning Outcomes

Syllabus 2013:

1. Probability and statistical inference

• Probability: including Bayes' rule, Markov chains and Monte Carlo integration.
• Parametric statistical models: including likelihood functions, estimators, and an introduction to Bayesian inference.

2. Supervised learning: classification and regression

• Introduction: training versus test data, notions of overfitting and cross-validation.
• Regression: including least squares, ridge regression and Bayesian approaches.
• Classification, including multivariate class-conditional distributions and generalised linear models.

3. Unsupervised learning: dimensionality reduction and clustering

• Dimensionality reduction, including principal components analysis
• Clustering, including k-means, mixture models and the expectation-maximisation algorithm

Bibliography:

Recommended texts:

Probability and statistics 

• Probability and statistics, M. H. De Groot & M. J. Schervish (Addison-Wesley, 2002)
• Statistical Inference (2nd Ed), Casella & Berger (Duxbury Press, 2001)
• Bayesian Data Analysis, A. Gelman et al. (Chapman & Hall, 2004)

Machine learning

• The Elements of Statistical Learning: Data Mining, Inference, and Prediction. 2nd Ed. (Springer, 2009)
• Pattern Classification , R. O. Duda et al. (Wiley, 2000)

Matlab

• Getting Started with MATLAB: A Quick Introduction for Scientists and Engineers (OUP, 2009)
• Mastering MATLAB 7, D. Hanselman & B. Littlefield (Pearson, 2012)

Teaching 2016:

Lectures per week	2 x 2 hours
Classwork sessions per week	Occasional Lecture: Mondays 14:00-16:00 Fridays 14:00-16:00 D1.07 Complexity
Module duration	10 weeks
Total contact hours	40
Private study and group working	60

Assessment information:

 

Week	Assessment	Issued	Deadline	%credit
1-4	Written assignment	5 Feb	19 Feb	25
4-8	Written assignment	26 Feb	11 Mar	25
10	Oral Examination	14 Mar		50