Dr Ben Goudey

WORKSHOP: An introduction to machine learning in life sciences

Machine learning promises to revolutionise life science research by speeding up data analysis, enabling prediction of biological patterns and modelling complex biological systems.

But what exactly is machine learning and when should you use it?

This hands-on online workshop provides a high-level introduction to machine learning: what it is, its advantages and disadvantages compared to traditional modelling approaches and the types of scenarios where it may be the right tool for the job.

Using example datasets and basic machine learning pipelines we contrast a few commonly used algorithms for constructing predictive models and explore some of their trade-offs. We discuss common pitfalls in how machine learning is applied and evaluated, with a focus on its application in the life-sciences and clinically settings, to help you recognise overly optimistic results. We discuss how and why such errors arise and strategies to avoid them.

Keywords: Machine learning, AI, predictive models

Requirements: A laptop will be required. It will be assumed that participants have some programming experience in R. No prior machine learning or statistical knowledge will be assumed. All analysis will use Google Colab which will required a Google account.

Relevance: This workshop will be most relevant to those looking to understand high level concepts in machine learning, either to apply them in their own work or want to understand the limitations in the work of others. It will also be of interest to ML practitioners who want to understand the sources of bias in life-sciences datasets and how to overcome them.