Scalable inference; statistical, algorithmic, computational aspects
SIN
3 July 2017 to 28 July 2017
Programme theme
The complexity and sheer size of modern data sets, of which ever increasingly demanding questions are posed, give rise to major challenges and opportunities for modern statistics. While likelihood-based statistical methods still provide the gold standard for statistical methodology, the applicability of existing likelihood methods to the most demanding of modern problems is currently limited. Thus traditional methodologies for numerical optimisation of likelihoods, and for simulating from complicated posterior distributions, such as Markov chain Monte Carlo and Sequential Monte Carlo algorithms often scale poorly with data size and model complexity, and thus fail for the most complex of modern problems.
The area of computational statistics is currently developing extremely rapidly, motivated by the challenges of the recent big data revolution, and enriched by new ideas from machine learning, multi-processor computing, probability and applied mathematical analysis. Motivation for this development comes from across the physical biological and social sciences, including physics, chemistry, astronomy, epidemiology, medicine, genetics, sociology, economics - in fact it is hard to find problems not enriched by big data and the resultant associated statistical challenges.
This programme will focus on methods associated with likelihood, its variants and approximations, taking advantage of, and creating new advances in statistical methodology. These advances have the potential to impact on all aspects of science and industry that rely on probabilistic models for learning from observational or experimental data.
Intractable likelihood problems are defined loosely as ones where the repeated evaluation of likelihood function (as required in standard algorithms for likelihood-based inference) is impossible or too computationally expensive to carry out. Scalable methods for carrying out statistical inference are loosely defined to be methods whose computational cost and statistical validity scale well with both model complexity and data size.
Understanding and developing scalable methods for intractable likelihood problems requires expertise across statistics, computer science, probability and numerical analysis. Thus it is imperative that the programme be broad, covering statistical, algorithmic and computational aspects of inference. The programme will cut across the traditional boundary between frequentist and Bayesian inference, and will incorporate both statistics and machine learning approaches to inference. Central to the focus will be the close integration of algorithm optimisation with the opportunities offered, and constraints imposed by modern multi-core technologies such as GPUs.
The first week of the programme will feature a broad-focused workshop, and more application specific activities will take place later.
Please note that the programme will also feature a one day workshop entitled "Sampling methods in statistical physics and Bayesian inference". This will take place on 18 July 2017 and attendance is free. To register your place please email samuel.livingstone[at]bristol.ac[dot]uk (statistics) or michael.faulkner[at]bristol.ac[dot]uk (physics). Further details below.
SIN programme: interview with participant Kerrie Mengersen
Workshops
Scalable statistical inference
3 July 2017 to 7 July 2017
Seminars
| Tuesday 18th July 2017 | |||
|---|---|---|---|
| 10:20 to 11:00 |
Joris Bierkens Delft University of Technology |
Room 1 | 
|
| 11:00 to 11:40 |
Robert Jack University of Bath |
Room 1 |
|
| 12:10 to 12:50 |
Michel Manon Orange Lab |
Room 1 |
|
| 12:50 to 13:30 |
Michela Ottobre Heriot-Watt University |
Room 1 |
|
| 15:00 to 15:40 |
Anthony Maggs City of Paris Industrial Physics and Chemistry Higher Educational Institution, (ESPCI) |
Room 1 |
|
| 15:40 to 16:20 |
Andrew Duncan Imperial College London |
Room 1 |
|
Organisers
- Christophe Andrieu University of Bristol
- Paul Fearnhead Lancaster University
- David Firth University of Warwick
- Chris Holmes University of Oxford
- Gareth Roberts University of Warwick
Speakers
- Joris Bierkens Delft University of Technology
- Andrew Duncan Imperial College London
- Robert Jack University of Bath
- Anthony Maggs City of Paris Industrial Physics and Chemistry Higher Educational Institution, (ESPCI)
- Michel Manon Orange Lab
- Michela Ottobre Heriot-Watt University
Participants
- Christophe Andrieu University of Bristol
- Louis Aslett University of Oxford
- Joris Bierkens Delft University of Technology
- Alexandre Bouchard University of British Columbia
- Nicolas Brosse École Polytechnique
- David Draper University of California
- Andrew Duncan Imperial College London
- Alain Durmus École Polytechnique
- Paul Fearnhead Lancaster University
- David Firth University of Warwick
- Michael Giles University of Oxford
- Peter Glynn Stanford University
- Chris Holmes University of Oxford
- Paul Jenkins University of Warwick
- Chris Jewell Lancaster University
- Adam Johansen University of Bristol
- Michael Jordan University of California, Berkeley
- Luke Kelly University of Oxford
- Krzysztof Latuszynski University of Warwick
- Anthony Lee University of Warwick
- Wentao Li Lancaster University
- Sam Livingstone University of Bristol
- Matthew Moores University of Warwick
- Eric François Moulines Télécom ParisTech
- Christopher Nemeth Lancaster University
- Helen Ogden University of Southampton
- Murray Pollock University of Warwick
- Nancy Reid University of Toronto
- Sylvia Richardson University of Cambridge
- Christian Robert CNRS & Université Paris-Dauphine
- Gareth Roberts University of Warwick
- Havard Rue Norwegian University of Science and Technology
- Steven Scott Google
- Chris Sherlock Lancaster University
- Alexandre Thiery National University of Singapore
- Cristiano Varin Università Cà Foscari di Venezia
- Dootika Vats University of Warwick; University of Minnesota
- Matti Vihola University of Jyväskylä
- Jessie Wu University of Oxford
- Sinan Yildirim Sabanci University
- Giacomo Zanella Bocconi University
Supported By
