School of Electronics and Computer Science:
COMP6036 Advanced Machine Learning
Basic Information
| School | |
|---|---|
| Known as | COMP6036. |
| Session and Semester | Semester Two, 2011 - 2012 |
| Credit | 20 Credit Points |
| Unit Leader | Prof Steve R Gunn |
| Moderators | Prof RI "Bob" Damper |
| Study | Reading and Assignments 176 hours |
| Assessment | Examination 50% Coursework 50% |
| Coursework | One research report and one technical report. |
| Teaching | 24 lectures/seminars |
| Prerequisites and Exclusions | Prerequisites: COMP3032 - Intelligent Algorithms. |
| Referral | On referral, this unit will be assessed 100% by examination. |
| Syllabus Approved |
Description
Aims
- To introduce key concepts in pattern recognition and machine learning; including specific algorithms for classification, regression, clustering and probabilistic modeling.
- To give a broad view of the general issues arising in the application of algorithms to analysing data, common terms used, and common errors made if applied incorrectly.
- To demonstrate a toolbox of techniques that can be immediately applied to real world problems, or used as a basis for future research into the topic.
Learning Outcomes
Knowledge and Understanding
Having successfully completed the module, you will be able to demonstrate knowledge and understanding of:
Having successfully completed the module, you will be able to demonstrate knowledge and understanding of:
- key concepts, tools and approaches for pattern recognition on complex data sets
- kernel methods for handling high dimensional and non-linear patterns
- state-of-the-art algorithms such as Support Vector Machines and Bayesian networks
- theoretical concepts and the motivations behind different learning frameworks
Intellectual Skills
Having successfully completed the module, you will be able to:
Having successfully completed the module, you will be able to:
- conceptually understand the role of pattern analysis and probabilistic modeling, together with the mathematical techniques this requires.
- analyse empirical results from a collection of methods on a data set in order to best estimate which will perform better in future independent tests.
- reason through the aspects of applying the techniques so that common pitfalls and misleading results are obtained through a simple naïve application.
Practical Skills
Having successfully completed the module, you will be able to:
Having successfully completed the module, you will be able to:
- analyse high-dimensional datasets using modern modeling techniques
- reason as to which method needs to be applied in a given situation depending on the specific application domain and additional requirements
Topics Covered
- Key concepts
- Supervised/Unsupervised Learning
- Loss functions and generalization
- Probability Theory
- Parametric vs Non-parametric methods
- Elements of Computational Learning Theory
- Kernel Methods for non-linear data
- Support Vector Machines
- Kernel Ridge Regression
- Structure Kernels
- Kernel PCA
- Latent Semantic Analysis
- Bayesian methods for using prior knowledge and data
- Bayesian inference
- Bayesian Belief Networks and Graphical models
- Probabilistic Latent Semantic Analysis
- The Expectation-Maximisation (EM) algorithm
- Gaussian Processes
- Key application areas
- Text classification and processing
- Problems in bioinformatics
Teaching and learning activities
Teaching methods include
- Lectures used to introduce material.
- Video lectures.
- Class-based discussion.
Learning activities include
- In-class exercises
- Group and individual based paper review, based on current research topics (coursework)
- Lab-based sessions with technical coursework to reinforce ideas
Methods of assessment
| Assessment method | Number | % contribution to final mark |
|---|---|---|
| Examination [exam] | 1 | 50 |
| Research Report [cwork] | 1 | 25 |
| Technical Report [cwork] | 1 | 25 |
Feedback and student support during module study
- Laboratory support for computing
- One-on-one support during drop-in clinics
- Interactive lectures to ensure student problems identified early
Relationship between the teaching, learning and assessment methods and the planned learning outcomes
- The lectures provide context, introduce students to available resources, including suggestions for student reading.
- The group and individual report allows students to demonstrate what they have achieved and learned from their studies.
- The technical courseworks give experience of applying the methods to real-world data.
- Exam ensures assimilation of the lectured material and ability to apply the ideas to novel problems.
Resources
Background Resources
- Christopher M. Bishop, Pattern Recognition and Machine Learning, Springer, 2006. [Library] [Shops]
- John Shawe-Taylor and Nello Cristianini, Kernel Methods for Pattern Analysis, Cambridge University Press, 2004. [Library] [Shops]
- David J C MacKay, Information Theory, Inference, and Learning Algorithms, Cambridge University Press, 2003. (Free screen readable version available at http://www.inference.phy.cam.ac.uk/mackay/itila/book.html) [Library] [Shops]
- David Barber, Bayesian Reasoning and Machine Learning, Cambridge University Press, 2011. (Free screen readable version available at http://www.cs.ucl.ac.uk/staff/d.barber/brml)
Taught to
COMP6036
Pt IV MEng Computer Science with Artificial Intelligence (Optional)Pt IV MEng Computer Science (Optional)
Pt IV MEng Computer Science with Distributed Systems & Networks (Optional)
Pt IV MEng Computer Science with Image and Multimedia Systems (Optional)
Pt IV MEng Computer Science with Mobile and Secure Systems (Optional)
MSc in Artificial Intelligence (Optional)
Non-existing cohort: "csMScCo" (Optional)
MSc in Software Engineering (Optional)
Pt IV MEng Electronic Engineering (Optional)
Pt IV MEng Electronic Engineering with Artificial Intelligence (Optional)
Pt IV MEng Electronic Engineering with Computer Systems (Optional)
Pt IV MEng Electronic Engineering with Nanotechnology (Optional)
Pt IV MEng Electronic Engineering with Optical Communications (Optional)
Pt IV MEng Electronic Engineering with Power Systems (Optional)
Pt IV MEng Electronic Engineering with Mobile and Secure Systems (Optional)
Pt IV MEng Electronic Engineering with Wireless Communications (Optional)
Pt IV MEng Software Engineering (Optional)
Students who are not registered on an ECS approved programme may take this module subject to meeting its pre-requisites and the availability of resources. To confirm this, please can you contact the module leader (as listed above) in the first instance. They will then refer you on to the appropriate director of studies for formal approval of your selection.