Machine Learning - CMS3503

Assessment 1: Other

Students will undertake a series of formative exercises in the practical sessions. Feedback, advice and guidance will be provided for these exercises.

Other

Assessment 2: Written Assignment

Development Task: Propose a software solution to the practical problem using machine learning method of your choice, which is to be documented in an evaluative report. Word count is indicative and does not include table of contents, tables, figures, diagrams, references and appendices.

Written Assignment

Assessment 1: Written Assignment

Investigation: Produce a scholarly report concerning the use of the machine learning methods in the application area you’ve selected. You should evaluate both the appropriateness and the readiness of a set of ML techniques to a problem class. Word count is indicative and does not include tabe of contents, tables, figures, diagrams, references and appendices.

Machine Learning techniques are now used widely in a range of applications either stand-alone or integrated with other AI techniques. The Machine Learning module allows you to obtain a fundamental understanding of the subject as a whole: how to embody machines with the ability to learn how to recognise, classify, decide, plan, revise, optimise etc. You will learn which machine learning techniques are appropriate for which learning problem, and what the advantages and disadvantages are for a range of ML techniques. We will consider the widely known data-driven approaches, and specific techniques such as “deep learning”, and investigate the typical applications and potential limitations of these approaches. We will introduce available tools and use them in practical classes, evaluating learning bias and characteristics of training sets. High profile applications of data driven, stand-alone, ML systems will be investigated, such as the AlphaGo method. Where data is sparse, and knowledge is already present in a system, we will investigate methods to improve heuristics of existing AI systems, and to learn or revise domain knowledge. This is essentially the area of model-driven ML, where is often integrated to other reasoning systems.

School of Computing and Engineering

https://library.hud.ac.uk/readinglists/CMS3503

Computing and Information Systems

CMS3503 - Machine Learning

Machine Learning

Critically evaluate the matching of machine learning techniques to application areas.

Apply the theoretical concepts and techniques to the solution of machine learning problems using appropriate ML tools.

Critically evaluate machine learning architectures for their appropriateness to applications.

Understand the fundamental principles and the theoretical concepts and techniques of Machine Learning.

Demonstrate a critical appreciation of current and emerging data-driven Machine Learning techniques and have a sound appreciation of their limitations.