Traditional statistical analysis and machine learning have many things in common but usually follow paths that are quite different. For prediction, traditional statistical analysis usually begins with a theory and a model and fits the parameters of the model to the data; machine learning follows a more pragmatic approach, allowing the data more freedom to prescribe the model.  Machine learning often leads to prognostic models that are more accurate but less interpretable.  For medical discovery, traditional statistical analysis usually begins by formulating a hypothesis and testing (the likelihood of) that hypothesis against the data; machine learning asks the data to formulate the (most likely) hypothesis. There are many synergies between these two disciplines and there is enormous potential in developing new fundamental theories and practical methods that transcend the boundaries of these disciplines leading to new and impactful methods to assist clinical practice and medical discovery. This workshop aims to foster such a dialogue and provide a forum for starting collaborations and for cross-fertilisation.

About the event


  • Learning and Inference from a Multitude of Data Sources
  • Dynamic Risk Prediction 
  • Early Disease Detection
  • Causal Inference and Individualised Treatment Effects
  • Rethinking Clinical Trials
  • Clinical Recommender Systems


Keynote speakers:

Oriol Vinyals (London DeepMind)

David Sontag (MIT Statistics and Data Science Center, USA)


Invited speakers:

David Alexander (University College London)

Jessica Barrett (University of Cambridge)

Franck Bretz (Novartis, Basel)

Sach Mukherjee (DZNE, Bonn)

Aditya Nori (Microsoft)

Blanca Rodriguez (University of Oxford)

Rajen Shah (University of Cambridge)

Sofia Villar (University of Cambridge)

Chris Williams (University of Edinburgh)

Chris Yau (University of Birmingham)

Serena Yeung (Harvard University)


Target audience:

Researchers, academics, postdocs, senior PhD students in statistics, machine learning and AI with an interest in healthcare and medicine.


Why unique?

Machine learning and statistics often offer solutions to the same types of problems, but have different merits and provide different advantages. We believe that there is scope for important theoretical and practical advances when these two disciplines join forces to assist clinical practice and medical discoveries. The two proponents are already working together to combine their expertise to develop the next generation of methods for medicine. This workshop aims to bring together these two communities by starting a friendly and deep dialog among them and fostering ideas for areas of collaboration.


Networking and competition:

There will be an interactive networking session at the end of day 1 (Monday 25 March) to encourage biostatisticians and machine learners to plan joint research. This will include refreshments and food and a competition for writing joint abstracts. Prizes for the joint abstract competition will be announced on day 2 of the workshop.



  • 20 x student places - £40 each
  • 80 x full places - £100 each

Lunch, refreshments and interactive networking session included in registration cost

Places are limited to 100 delegates.



Day 1: 9:30 - 19:30

Day 2: 8:30 - 16:15


52.176333194762, 0.14279007911682