The Artificial Intelligence & Machine Learning Research Group in the Department of Computer Science cordially invites all University of Bath staff and students to an AI Talk with Neill Campbell entitled ‘No Free Lunches in Machine Learning’. This is taking place on the 28th February at 13.15pm-14:05pm in the Chancellor’s Building, CB 1.11. Register here for the talk.
Abstract
With the media attention around Machine Learning (a particular branch of Artificial Intelligence), it can be difficult to build reasonable intuitions or expectations around what such systems should deliver or how we can incorporate “machine learning technology” into a particular field; this could be as academic researchers or more broadly across society and different industries.
This talk will provide an accessible introduction to some foundational aspects of machine learning that can be used to establish some reasonable expectations for how machine learning may (or may not!) be helpful to a particular application. There will be discussion of the philosophy behind some approaches and a gentle introduction to technical aspects; the intention is to avoid assuming any particular mathematical or computational background knowledge.
Points considered during the talk will include:
- Are all datasets equal or how to choose your data?
- Gotchas in Machine Learning: What ML can and can’t do for you.
- Average vs Worst Case Machine Learning
- Machine Learning and Causality
- Trade-offs in Machine Learning
Bio
Neill Campbell is Professor of Visual Computing and Machine Learning in the Department of Computer Science at the University of Bath. His main area of research involves learning models of shape (2D and 3D), appearance and dynamics from images and video. In particular, he is interested in performing this in an automatic or interactive fashion that allows these technologies to be put to use in a variety of applications without requiring users to have computer vision or graphics expertise. He is also interested in machine learning problems where data are scarce or expensive to obtain (e.g. annotations from expert clinicians) and when uncertainty in the resulting output is important (e.g. medical applications).
Neill is the Director of the Centre for the Analysis of Motion, Entertainment Research and Applications (CAMERA) [1] that researches and applies visual computing and machine learning technology in the fields of Entertainment, Health and Sports Science. He is also co-director of the Centre for Mathematics and Algorithms for Data [2], an inter-disciplinary group between Mathematical Sciences and Computer Science that studies, in particular, the theoretical underpinnings of Machine Learning and Data Science. He has been at Bath since 2014, previously working as a Research Associate in Computer Science at University College London after a PhD in Engineering at the University of Cambridge.
This is the fourth talk in the “AI Talks” series, an in-depth exploration of current topics in AI that is accessible to a broad audience. For more information about the talks and to see future talks, please see https://www.bath.ac.uk/announcements/ai-talks/.
