Probabilistic Machine Learning Takes Centre Stage at the ELLIS Unit Cambridge Summer School 2026
Hosted across the University of Cambridge Department of Engineering and Pembroke College, the event was organised by the ELLIS Unit Cambridge, which brings together researchers working at the forefront of machine learning and AI to foster collaboration, knowledge exchange, and the development of responsible AI technologies.
The programme welcomed 122 attendees and featured 21 speakers from across the machine learning and artificial intelligence community. Each day focused on a distinct theme, bringing together experts and participants around key areas of modern AI research: Introduction to Probabilistic Modelling; Uncertainty in Deep Learning and Uncertainty Quantification; Generative Models; Decision Making Under Uncertainty; and Probabilistic Methods in the Sciences. Through lectures, discussions, and panel sessions, the Summer School explored both the theoretical foundations and practical applications of Probabilistic Machine Learning, highlighting the importance of uncertainty-aware approaches in the development of reliable and interpretable AI systems.
The academic programme featured contributions from leading researchers working across the breadth of Probabilistic Machine Learning. The opening day on Introduction to Probabilistic Modelling featured talks from Carl Henrick Ek, Neill Campbell, Carl Rasmussen, Victor Picheny, and Dragos Margineantu, who explored foundational approaches to probabilistic modelling and its applications. The second day, focused on Uncertainty in Deep Learning and Uncertainty Quantification, brought together Gianni Franchi, Laurence Aitchison, Samuel Müller, and Theodore Papamarkou, alongside a panel discussion addressing challenges and opportunities in uncertainty-aware AI.
The programme continued with sessions on Generative Models, featuring Jes Frellsen, Alex Tong, Adhithya Saravanan, Alexander Denker, and Henry Moss, before moving to Decision Making Under Uncertainty, where Jasmine Bayrooti, Disha Hegde, and Sylvie Delacroix contributed perspectives on decision-making approaches in uncertain environments. The final day, dedicated to Probabilistic Methods in the Sciences, featured talks from Sophia Tang, Will Baldwin, Marwin Segler, and Neil Lawrence, highlighting the growing role of probabilistic methods in scientific research and discovery.
The Summer School concluded with a poster session, providing attendees with the opportunity to present and discuss their own research. This final session encouraged further exchange between participants and speakers, strengthening the collaborative atmosphere that characterised the week.
The Summer School was free to attend, although a £500 sustainability fee was charged to participants travelling to the UK by air, reflecting the organisers’ commitment to environmentally responsible events and reducing the carbon impact of international academic gatherings.
Alongside the academic programme, attendees enjoyed a range of social activities designed to encourage networking and community building. These included a formal dinner at Sidney Sussex College, a science-themed walking tour of Cambridge, and an evening at the Cambridge Shakespeare Festival, set within the historic grounds of Cambridge colleges.
To help foster connections throughout the week, the organisers created a WhatsApp group that enabled attendees to arrange informal dinners, punting on the River Cam, social gatherings, and even group viewings of World Cup matches at local pubs. These activities helped create a welcoming and collaborative atmosphere that extended beyond the lecture theatre, allowing participants to build new connections and continue discussions throughout the week.
Picture credits:
Punting: Christoph Griesbacher
Tour: Leander Fritsch
Other: Majid Liaquat
Text written with contributions by Kimberly Cole:
