Topics in Machine Learning

COMP64501 Topics in Machine Learning 15 Semester 1 Postgraduate Taught Level 6 Mauricio Alvarez Lopez Unit coordinator 1 FHEQ level (Framework for Higher Education Qualifications) ' Masters/Integrated Masters P4 ' European Credit Transfer & Accumulation System Rating : 7.5 Machine Learning (ML) is a subfield of AI that focuses on fitting mathematical functions to data for practical applications such as predictive modelling. It is concerned with creating mathematical "data structures" that allow a computer to exhibit behaviour that would normally be considered human. Typical applications include spam filtering, speech recognition, medical diagnosis and weather prediction. The data structures we use (known as "models") come in various forms, e.g. trees, graphs, algebraic equations and probability distributions. The emphasis is on constructing these models automatically from data – for example, making a weather predictor from a data file of historical weather patterns. Machine Learning (ML) is a subfield of AI that focuses on fitting mathematical functions to data for practical applications such as predictive modelling. It is concerned with creating mathematical "data structures" that allow a computer to exhibit behaviour that would normally be considered human. Typical applications include spam filtering, speech recognition, medical diagnosis and weather prediction. The data structures we use (known as "models") come in various forms, e.g. trees, graphs, algebraic equations and probability distributions. The emphasis is on constructing these models automatically from data – for example, making a weather predictor from a data file of historical weather patterns. This module introduces students to supervised machine learning and deepens their knowledge by discussing state-of-the-art deep learning technologies. It covers a range of ML tasks and models and the associated mathematical concepts. The unit starts with an introduction to the basic definitions of machine learning, and the stages of an end-to-end Machine Learning project. The first mathematical functions studied are linear predictive models followed by neural networks, convolutional neural networks and deep versions of these models. The module also studies technologies for sequence modelling including language models, recurrent neural networks and transformers. Machine learning is now likely the most prominent branch of Artificial Intelligence (AI) and currently underpins the wide societal interest in AI as a general-purpose technology. This course aims to introduce the concepts behind various Machine Learning techniques, including how they work, and to use existing software packages to illustrate how they behave. 1. Describe the fundamental concepts of machine learning.2. Describe the advantages and disadvantages of linear and non-linear approaches to supervised learning.3. Explain the models and algorithms involved in a pipeline for supervised learning.4. Use basic concepts of linear algebra to write scalar objective functions in terms of vector and matrix operations.5. Design, develop, and evaluate specific predictive models for data-driven applications using tools such as Scikit-learn and automatic differentiation frameworks.6. Critically assess technologies and analyse their suitability for specific application scenarios. Analytical skills Innovation/creativity Project management Problem solving Research Written communication 1. Introduction to Machine Learning2. An end-to-end Machine Learning project3. Linear models for regression and classification4. Neural Networks and Convolutional Neural Networks5. Sequential models: Language models, Recurrent Neural Networks and Transformers. - Asynchronous material in the form of video lectures, formative exercises, lecture slides and code examples delivered via the virtual learning environment, lectures and supported tutorials.- Weekly lectures consolidating asynchronous materials and providing opportunities for discussion and questions.- Bi-weekly supervised lab sessions and tutorials that provide support for coursework and formative exercises. 1 Written exam 80% 8 Practical skills assessment 20% Individual and general feedback for the Practical skills assessment.Formative assessment in Lectures and Practical sessions. Introductory courses on Linear Algebra, Probability, Calculus and Python Programming. Y Simon J.D. Prince, Understanding Deep Learning, MIT Press, 2023.Christopher Bishop, Deep Learning: Foundations and Concepts, Springer, 2023.Aurélien Géron, Hands-On Machine Learning with Scikit-Learn, Keras and Tensor-Flow, O’Reilly, 3rd Edition, 2022.Moritz Hardt and Benjamin Recht, Patterns, Predictions, and Actions: Foundations of Machine Learning, Princeton University Press, 2022.Kevin Murphy, Probabilistic Machine Learning: an Introduction, First edition, The MIT Press, 2022.Simon Rogers and Mark Girolami, A First Course in Machine Learning, Chapman and Hall/CRC Press, 2nd Edition, 2016.Christopher Bishop, Pattern Recognition and Machine Learning, Springer, 2007. Assessment written exam 2 Lectures 15 Practical classes & workshops 10 Tutorials 5 0 118