skip to content

MPhil in Machine Learning and Machine Intelligence

 

The 2021-2022 course is based on a set of core modules.  These modules will provide an introduction to the field as well as a good understanding of advanced techniques, prepare

ing students for their later research.  

Students must also choose two elective modules. One or both of these must be selected from two core electives: Computer Vision or Natural Language Processing.

Students may choose their second elective module from the 4th Year undergraduate programme.  
 

Below are the core modules planned for 2021-2022  (Please note that the course reserves the right to change or withdraw modules).

 

Introduction to Machine Learning (MLMI 1), Michaelmas Term

Principal Lecturer: Professor Richard Turner

On completion of this module, students should understand:

  • Basic machine learning concepts concerning probabilistic modelling and fitting
  • Approaches to regression and classification
  • Methods for clustering and time series modelling
  • Parameter estimation techniques including maximum likelihood fitting, expectation maximisation, and dynamic programming
Deep Learning and Structured Data (4F10), Michaelmas Term

This module aims to teach the basic concepts of deep learning and forms of structure that can be used for generative and discriminative models. In addition the use of models for classifying structured data, such as speech and language, will be discussed

As specific objectives, by the end of the course students should be able to:

  • Understand the basic principles of pattern classification and deep learning;
  • Understand generative and discriminative models for structured data;
  • Understand the application of deep-learning to structured data;
  • Be able to apply pattern processing techniques to practical applications.
Probabilistic Machine Learning (4F13), Michaelmas Term

The exposition of the material will be centered around three specific machine learning areas:

  • supervised non-parametric probabilistic inference using Gaussian processes
  • the TrueSkill ranking system
  • the latent Dirichlet Allocation model for unsupervised learning in text.
Speech Recognition (MLMI 2), Michaelmas Term
Computer Vision (MLMI 12), Michaelmas Term
Natural Language Processing (MLMI 13), Michaelmas Term
Advanced Machine Learning (MLMI 4), Lent Term

The aim of this module is to teach advanced topics that will enable students to follow state-of-the-art research in machine learning.  On completion students should:

  • understand advanced topics in machine learning
  • be able to read current research papers in the field
  • be able to implement state of the art learning algorithms
  • be ready to conduct research in the field.
Reinforcement Learning and Decision Making (MLMI 7), Lent Term

This module introduces basic principles of sequential decision making under uncertainty and the application in Reinforcement Learning and Control. Foundations and recent algorithms are covered.  On completion, students should understand:

  • the foundations of sequential decision making and reinforcement learning
  • the connections between control and reinforcement learning
  • the exploration vs exploitation trade-off

 

 

Neural Machine Translation and Dialogue Systems (MLMI 8), Lent Term

Principal lecturer: Prof Bill Byrne

The aim of this module is to provide an introduction to machine translation and task-oriented dialogue systems as problems that can be addressed by machine learning.   The presentation will employ sequence-to-sequence models to develop a uniform approach to these problems.

 

Designing Intelligent Interactive Systems (MLMI 10), Lent Term

The aims of this module are to:
• provide a basic understanding of design and human-computer interaction theories and methods
• introduce a systematic process for designing intelligent interactive systems
• introduce design tactics for realising effective intelligent interactive systems.

As specific objectives, by the end of the module students should be able to:
• apply a systematic design engineering process to design an interactive system.
• apply qualitative and quantitative methods to gain an understanding of users’ needs and wants.
• understand elementary human behavioural theory, as it applies to user interface design.
• model user behaviour and understand the limitations and implications of such modelling.
• understand how user interfaces can be optimised and have some knowledge of common optimisation approaches.
• understand design strategies for interactive systems that infer or predict user behaviour.
• understand the role of verification and validation and have basic knowledge of common verification and validation strategies for interactive systems.

Spoken Language Processing and Generation (MLMI 14)

Details to be advised.