Program

The SnB program consists not only of invited and contributed plenary sessions, but also pre-conference courses, poster session and many opportunities to network. SnB offers a great opportunity for statisticians in academia, industry, and government to exchange ideas and explore opportunities for collaboration.

Check back soon for a description of pre-conference courses and to get the full SnB program. Meanwhile please find below the scientific topics that will be adressed:
  • Adaptative trials: from Bayesian approaches to reinforcement learning
  • Open, replicable, understandable science
  • Regulatory point of view
  • The data in the data science: integrating heterogeneous data & generating
    synthetic data
  • Causality, causal inference beyond randomization
  • Flexible modelling, model uncertainty, model specifications
  • High dimensional data
  • Lessons learned from Covid19, Accelerated drug development



Conference Program Overview


Time


19-SEP-2022


20-SEP-2022


21-SEP-2022


Morning


Short Course
Introduction to Neural Networks and their properties


Plenary 2
Statistics and Machine Learning: friends or foes?


Discussion/Debate session

Lunch Break Lunch Break Lunch Break

Afternoon


Plenary 1
Big data to the rescue of drug development challenges?


Plenary 3
Getting medicines to patients faster – the role of innovative designs


Plenary 4
Lessons learnt from the COVID-19 experience


Evening


Poster & Wine Session


Conference Dinner


Confirmed invited speakers :
  • Gary Collins
  • Harald Binder
  • Marc Buyse
  • Chris Holmes
  • Emilie Kaufmann
  • Stephen Senn
  • Ewout Steyerberg
  • Wolfgang Jacquet
  • Michal Abrahamowicz
The preliminary program is available here.

Short course

Course description
The 3-hour course will provide an overview of the research on neural networks and deep learning.
The essential components of the design and learning of modern neural architectures will be introduced.
The original ideas that make deep learning successful today will be identify as well as the difficulties related to the development of these models.
Finally, there will be addressed the issues concerning their functioning, their practice and the perspectives to push the limits of their performance and the spectrum of their applications.
Participants are encouraged to bring a laptop with R installed to be able to make the most of this training.


Trainer: Thierry Artières
Thierry Artières Thierry Artières is a professor in Computer Science at the Ecole Centrale Marseille. He leads the QARMA's team of Machine Learning in Marseille at the Computer Science and Systems Laboratory (LIS - UMR 7020) of Aix Marseille University. He is the author or co-author of around 100 publications in conferences and journals. He conducts research in machine learning, deep learning and representation learning, in particular in neuroscience and particle physics.