KL7016 - Networks and Machine Learning

You will learn through a series of lectures, seminars and practical problem-solving sessions, which include classroom discussions and presentations. Seminars will be scheduled at regular intervals to allow exploration of the theoretical background to the techniques covered in the lectures as well as attempt the practical analysis of selected problems. Lectures allow you develop relevant theoretical aspects underpinning network theory, statistics and machine learning and understand how to apply the techniques to examples and unseen problems.

Formative feedback is available in classes as you get to grips with new techniques and solve problems. In addition, we operate an open-door policy where you can meet in person and/or online with your module tutor to seek further advice or help if required. Your ability to use the relevant theory to identify and evaluate solutions to set problems is assessed by a coursework and a group workl.

General and individual feedback on the assessment will be given via eLearning Portal/Blackboard. An opportunity to discuss work further will be available on an individual basis when work is returned and also through the open-door policy.

Direct contact with the module team during the lectures and workshops will involve participation in both general class discussions as well as one to one discussion during problem-solving workshops. This gives you a chance to get immediate feedback pertinent to your particular needs in a session. Further feedback and discussion with the module team are also available at any time through our open-door policy. In addition, all teaching materials including relevant research articles are made available through the e-learning portal.

All modules at Northumbria include a range of reading materials that students are expected to engage with. Online reading lists (provided after enrolment) give you access to your reading material for your modules. The Library works in partnership with your module tutors to ensure you have access to the material that you need.

Knowledge & Understanding:
1. Analyse and use of algorithms for network structures
2. Principles and applications of complex networks for supervised and unsupervised machine learning

Intellectual / Professional skills & abilities:
3. Develop efficient solutions and algorithms for advanced problems in machine learning

Personal Values Attributes (Global / Cultural awareness, Ethics, Curiosity):
4. Manage your own learning, through knowledge of available reading sources, including advanced texts and research papers.
5. Effectively and concisely communicate complex ideas in written form.

SUMMATIVE
1. Coursework (30%) 1,2,3
(Assignment including open ended questions and problems – wordcount: max 1000 words + derivations + codes + graphs + tables + plots)
2. Presentation 70%) – 1,2, 3, 4, 5
(Individual project presentation with the use of electronic slides and discussion, 15min presentation + 10min discussion)

FORMATIVE – 1, 2, 3, 4, 5
Formative assessment will be available during problem-solving workshops through lecturer-student interactions and discussions around the set questions, allowing students to extend, consolidate and evaluate their knowledge.

Formative feedback will be provided on student work and errors in understanding will be addressed reactively using individual discussion. Solutions to problems will be provided after the students have attempted the questions, allowing students to receive feedback on the correctness of their solutions and to seek help if matters are still not clear.

None

None

Theory of networks provide a universal paradigm to understand, describe and build complex and intelligent systems. Network structures spontaneously arise in nature and network models are developed to study a variety of phenomena in e.g. physical sciences, biology, social science. Complex network structures are built to support existing technologies and create new ones, from transportation networks to the internet, from the world wide web to the internet of things. The combination of network theory and statistics leads to machine learning, the science of making computers act without being explicitly programmed, opening a potential new route towards human level Artificial Intelligence.

This module will provide you with the theoretical concepts to model large complex and random systems as well as the theoretical underpinning of machine learning and the specific application to artificial intelligence. You will learn how to formulate mathematical models for systems with a large number of interacting components, provide quantitative description of their behaviour and performance and present the results. For the study of models and their applications you will develop analytical and numerical techniques using professional software and languages.

You will learn through a combination of lectures, seminars and problem-solving and computer-based sessions. Problem-solving sessions will be an opportunity to address open research problems; they will often address topics with links beyond the discipline, strengthening your transferable skills and employability.

The module is assessed through a coursework and a groupwork presentation, which will cover all aspects of the module and will assess your problem-solving abilities when applied to unseen problems.