Demystifying Machine Learning: Theory and Applications

Considering the impact of machine learning on a personal and industrial level 

Module details

  • Offered to 3rd Year students in Spring term
  • Thursdays 16:00-18:00
  • Planned delivery: On campus (South Kensington) & Online
  • 1-term module worth 5 ECTS  
  • Available to eligible students as part of I-Explore  

What is Machine Learning really? And how can it be applicable to you and others around you? This module aims to provide you with an intuitive understanding of the foundations of Machine Learning and how it can be applied to different disciplines. Besides learning the theory behind basic Machine Learning concepts, you will also apply what you have learnt to identify a user problem that can potentially be solved using Machine Learning and deliver a proof of concept of your proposed solution. By the end of the module, you will be able to cut through the Machine Learning hype in the mass media and explain to others what Machine Learning really is all about! 

You will learn about the dynamics of machine learning and the different learning systems that this technology employs. You will look at this topic through two lenses; theory and application. You will work both independently and in a group setting and utilize your presentation and research skills. 

Please note: The information on this module description is indicative. The module may undergo minor modifications before the start of next academic year. 

Accordian

Learning outcomes

By the end of this module, you will better be able to: 

  • Describe core machine learning principles, models and algorithms 
  • Explain basic machine learning concepts to different audiences, such as laypeople, domain experts, and those with a more technical background 
  • Identify problems and opportunities in multiple disciplines, including your own, that might benefit from machine learning 
  • Apply machine learning principles and algorithms to solve a problem in an applied discipline, and critically evaluate other people’s solutions 
  • Reflect on how machine learning is applicable to you at a personal level, within your surrounding context and the wider context 

Indicative core content

The module will cover theoretical foundations of machine learning concepts, models, and algorithms. You will aim to gain an intuitive understanding of topics such as: 

  • What machine learning is all about 
  • The importance of data and data pre-processing in machine learning 
  • Evaluating machine learning systems 
  • Models and algorithms for supervised learning, such as linear models (linear and logistic regression) and non-linear models (e.g. neural networks) 
  • Models and algorithms for unsupervised learning problems like clustering and density estimation 

You will also apply your knowledge to gain hands-on experience in solving problems in a specific discipline.

Learning and teaching approach

The course will run in two parallel 'threads': theory and application. Both threads will run throughout the whole term. 

 The 'theory' thread will focus on the technical aspects of machine learning. These will be delivered asynchronously via a mix of prepared videos and materials for self-guided study and more advanced independent study. There will also be guest lectures from different domains or related topics - these may be delivered synchronously or asynchronously subject to the availability of guest speakers. 

The 'application' thread will focus on the practical aspects of machine learning, where you apply machine learning to solve problems in a specific discipline. These will be done via a group project, where you will identify a user problem in a specific domain that can potentially be solved using machine learning, propose a solution, and come up with a proof of concept for your proposed solution. Tutors/mentors will guide you towards developing your solution. 

 For both threads, you will be demonstrating your understanding by trying to communicate machine learning concepts to different types of audiences, including laypeople and people with a more technical background.

You will receive immediate verbal feedback during any live assessments (the group projects). Marks from peer assessments will be averaged, and written qualitative feedback from peers and/or the teaching team will be returned to you once compiled. 

Assessment

Coursework: 

  • Self-reflection document (10%) 

Practical: 

  • Project Milestone 1: Pitch (15%) 
  • Project Milestone 2: Promotional material (25%)  
  • Project Milestone 3: Proof of concept (50%)  

Key information

  • Requirements: It is compulsory to take an I-Explore module during your degree (you’ll take an I-Explore module in either your 2nd or 3rd year, depending on your department). You are expected to attend all classes and undertake approximately 105 hours of independent study in total during the module. Independent study includes for example reading and preparation for classes, researching and writing coursework assignments, project work and preparing for other assessments 
  • I-Explore modules are worth 5 ECTS credit towards your degree; to receive these you will have to pass the module. The numerical mark that you obtain will not be included in the calculation of your final degree result, but it will appear on your transcript 
  • This module is designed as an undergraduate Level 6 course 
  • This module is offered by the Department of Computing