Tutor: Dr John Pinney
Course Level: Level 1
Prerequisites:  No prior experience of programming is required.
Class Duration: 3 x 2 hour sessions
Format: Teams session with live teaching and hands-on practice

Description:
Machine learning is a broad topic, with a wide range of applications in scientific research. 

In this series of lectures, we will introduce the fundamental concepts of unsupervised and supervised learning, including the training, testing and evaluation of models for classification and regression.  We also explore the basic theory of neural networks and discuss their applications to deep learning.

 Examples will be provided using Orange data science environment.  No prior experience of programming is required.

Syllabus:

  • Unsupervised learning
    • Dimensionality reduction
    • Clustering
  • Supervised learning
    • Generalised linear models
    • Logistic regression
    • Support-vector machines
    • Decision trees
  • Evaluating performance
    • Cross-validation
    • ROC curves
  • Improving performance
    • Feature selection
    • Ensemble methods
  • Artificial neural networks
    • Multi-layer perceptron
    • Stochastic gradient descent
    • Activation functions
    • Deep learning applications

Materials and pre-course set-up

Learning Outcomes:

After completing this workshop, you will be better able to:

  • Explain the difference between supervised and unsupervised learning.
  • Select a suitable machine learning method for a given application.
  • Prepare your own training and testing data sets.
  • Evaluate the performance of a machine learning experiment.

Dates

BOOK

Tuesday 18 May 2021, 14:00-16:00 (Part One)
Wednesday 19 May 2021, 14:00-16:00 (Part Two)
Thursday 20 May 2021, 14:00-16:00 (Part Three)
FULLY BOOKED, waiting list spaces available
Microsoft Teams

BOOK

Monday 14 June 2021, 10:00-12:00 (Part One)
Wednesday 16 June 2021, 10:00-12:00 (Part Two)
Friday 18 June 2021, 10:00-12:00 (Part Three)
FULLY BOOKED, waiting list spaces available
Microsoft Teams

BOOK

 

Summary of the table's contents

Students must attend all parts to be awarded the course credit