• Dublin City University logo

Machine Learning with Neural Networks: Foundations of AI

This course will cover learning a function from exemplars, Single-layer Neural Networks and Multi-layer Neural Networks.

  • Duration

    4 weeks
  • Weekly study

    0 hours

Learn about Single-layer Neural Networks and Multi-layer Neural Networks

This course will cover learning a function from exemplars, Single-layer Neural Networks, Multi-layer Neural Networks and activation functions.

It will also deal with the Back-propagation algorithm, Network design issues, supervised learning schemes, as well as other types of learning, Deep learning and character recognition.

What topics will you cover?

  • Topic 1: Machine Learning with Neural Networks. We will explain the motivation for machines like neural networks. We will contrast them with other types of prediction machines. We will then introduce the single layer neural network and explain its limitations. A demo on Ancient Brain is provided, with a coding exercise.
  • Topic 2: Multi-layer Neural Networks. Multi-layer Neural Networks. We explain how a multi-layer network is a “universal approximator”. We explain in detail the back propagation algorithm. We explain that it is a heuristic, and like any heuristic, needs tweaking and may sometimes fail.
  • Topic 3: Supervised learning in practice. Neural networks need a lot of minding. There are many design decisions to be made. If the design is wrong, the network cannot learn. There is often quite a search for a design that works. We have a detailed section explaining how to get neural networks working in practice.
  • Topic 4: Neural Networks (finish). We explain the place of supervised learning in relation to other forms of learning. We consider the new revolution in “deep learning” and exactly why neural networks have become so much better in modern times. Finally, we introduce the most challenging demo and exercise on Ancient Brain, the Character recognition neural network. This is an advanced demo that runs in the browser. Students are set a big practical to modify its code to make it better. This is the section where you can really shine. The course ends on a high note, with an algorithm that can really do something that is definitely “AI”.

Learning on this course

On every step of the course you can meet other learners, share your ideas and join in with active discussions in the comments.

What will you achieve?

By the end of the course, you‘ll be able to...

  • Demonstrate an understanding of the spectrum of neural networks, from perceptrons to multi-layer nets to deep learning.
  • Demonstrate an understanding of the difficulty of getting networks to work in practice. Networks need hand holding, or the learning will not happen.
  • Demonstrate the ability to code and modify a series of neural networks, from a simple perceptron up to a Character recognition neural network.
  • Explain that the coding on this course is in JavaScript in the browser, but you will be able to apply the concepts to other platforms.

Who is the course for?

The course is aimed at IT professionals in employment in Republic of Ireland registered companies. To qualify for direct entry they must have a Level 8 Honours Degree (2.2) or higher in Computer Science, Computing, Computer Applications or a related discipline. Applicants without these entry requirements (e.g., Level 7 degree or lower than an Honours 2.2 in a Level 8 degree) may be considered if they can demonstrate previously obtained competence equivalent to the entry requirements.

Who will you learn with?

Dr. Mark Humphrys is a lecturer at DCU. He has a BSc from UCD and a PhD from Cambridge. His research interests are in AI. He is the inventor of the coding site "Ancient Brain".

Who developed the course?

Dublin City University

Dublin City University is a young, dynamic and ambitious Irish university with a distinctive mission to transform lives and societies through education, research and innovation.

Endorsers and supporters

funded by

Skillnet Ireland

Learning on FutureLearn

Your learning, your rules

  • Courses are split into weeks, activities, and steps to help you keep track of your learning
  • Learn through a mix of bite-sized videos, long- and short-form articles, audio, and practical activities
  • Stay motivated by using the Progress page to keep track of your step completion and assessment scores

Join a global classroom

  • Experience the power of social learning, and get inspired by an international network of learners
  • Share ideas with your peers and course educators on every step of the course
  • Join the conversation by reading, @ing, liking, bookmarking, and replying to comments from others

Map your progress

  • As you work through the course, use notifications and the Progress page to guide your learning
  • Whenever you’re ready, mark each step as complete, you’re in control

Want to know more about learning on FutureLearn? Using FutureLearn