01 NOV


Machine Learning Theme Launch

Come and join us at the University of Leeds for the launch of our Machine Learning Theme.

Nexus, University of Leeds
1 Nov 2022 9:30 a.m. — 4 p.m.

Registration for presentation abstracts has now closed. Registration is now full, but please add your name to the waitlist on Eventbrite.

Open to all researchers, PGRs and anyone involved in Machine Learning at an N8 institution, this event will bring the Machine Learning community from across N8 together to find out more about N8 CIR, plans for the theme and all-important networking opportunities.

The event will have three key parts:

  • An introduction to N8 CIR and the use of Bede, N8 CIR's high-performance computing platform, for Machine Learning
  • Presentations and posters featuring Machine Learning research from the community
  • An opportunity for the community to hear about future plans and community building for the theme and to input their thoughts and ideas


9.30am – Welcome and housekeeping

Dr Serge Sharoff, School of Languages, Cultures and Societies, University of Leeds

9.40am – 10am – Introduction to N8 CIR

Dr Gillian Sinclair, N8 CIR Programme Manager

10am – 10.20am – Introduction to N8 CIR special interest groups

Dr Marion Weinzierl, N8 CIR RSE Theme Lead & Dr Peter Crowther, N8 CIR RIE Theme Lead

10.20am – 10.50am – Coffee

10.50am – 12:30pm – 5 talks

View abstracts

Parallel, efficient MCMC with Transport Elliptical Slice Sampling

Alberto Cabezas Gonzalez, University of Lancaster

Bayesian methods for extrapolating adult clinical trial data to improve the estimation of the effect of medical treatments in children.

Ruth Walker, University of York

Bayesian Deep Learning with Physics-informed Gaussian Processes

Thomas McDonald, University of Manchester

Visual bias mitigation driven by Bayesian epistemic uncertainties

Rebecca Stone, University of Leeds

The premise of approximate MCMC in Bayesian deep learning

Theodore Papamarkou, University of Manchester

12:30pm – 2pm – Lunch and posters

2pm – 2.40pm – 6 talks

View abstracts

A Bayesian Hierarchical Multifidelity Model for High-fidelity Predictions of Turbulent Flows

Saleh Rezaeiravesh, University of Manchester

Integrating autoencoder and Bayesian methods for batch process soft-sensor design

Sam Kay, University of Manchester

Autonomous Optimisation for Multistep Chemical Synthesis

Adam D. Clayton, University of Leeds

Toward Predicting Process Outcomes in Different Solvents: Solubility and Reactivity

Bao Nguyen, University of Leeds

Machine learning directed multi-objective optimization of mixed variable chemical systems

Oliver Kershaw University of Leeds

2.40pm – 3pm – Bede, the N8 CIR GPU machine, and Machine Learning

Dr Alan Real, N8 CIR Technical Director

3pm – 3.20pm – Coffee

3.20pm – 3.40pm – Machine Learning Workshop Feedback

Dr Marion Weinzierl, N8 CIR RSE Theme Lead & Dr Peter Crowther, N8 CIR RIE Theme Lead

3.40pm – 4pm - Future Plans for the Machine Learning Theme

Dr Gillian Sinclair, N8 CIR Programme Manager & Grainne Wrigley, N8 CIR Marketing and Communications Officer

Call for Presentation Abstracts (now closed)

Bayesian Methods in Machine Learning

Machine learning in any area requires data, both for training prediction models and for applying them to solve real-world problems. Bayesian methods are known to be useful to make sense of data, to make better models and to interpret the predictions, especially in the presence of noise in either input data or inference data. With this N8 CIR event, we want to share current practices in using Bayesian methods across a wide range of disciplines and applications, so that this can provide cross-pollination and enrich our understanding of which methods might be applicable.

Submit your 300 word abstracts to enquiries@n8cir.org.uk by the 30th of September. Successful presenters will be informed by mid-October.

Call for Posters

Poster submissions are welcome on any aspect of Machine Learning, simply provide a poster title when you register.

Return to event index