Previous Seminars

data

Markus Heinonen: Differential equations and deep learning

Differential equations describe the evolution of a system’s state, and are widely applied in natural sciences. Deep learning, on the other hand, is based on learning a sequential of the input...

Learn More
data

Pearse Keane: Reinventing the Eye Examination with Deep Learning

Ophthalmology is among the most technology-driven of the all the medical specialties, with treatments utilizing high-spec medical lasers and advanced microsurgical techniques, and diagnostics...

Learn More
data

Vikas Garg: Generalization and Representational Limits of Graph Neural Networks

Graphs provide a natural abstraction to model relational and strategic data in domains as diverse as biology (e.g., molecules), multiagent settings (e.g., online vendors on ecommerce platforms)...

Learn More
data

Neil Lawrence: Deploying Machine Learning: Intellectual Debt and AutoAI

Speaker: Neil Lawrence (University of Cambridge) Title: Deploying Machine Learning: Intellectual Debt and AutoAI Abstract: From the dawn of cybernetics, and across the last eight decades, we’ve worked

Learn More
data science image 2

Barbara McGillivray: Data science and historical texts: modelling meaning change from Ancient Greek to web archives

Over time, new words enter the language, others become obsolete, and existing words acquire new meanings. In Old English thing meant ‘a public assembly’ and now means more generally ‘entity’; chill...

Learn More
data

Naoaki Okazaki: Controllable headline generation

Headline generation is a special type of abstractive summarisation that generates a very short summary (headline) from a news article. Headline generation is reaching a practical level in terms of...

Learn More