Entrance hall of the ICMS Workshop

MCMC and particle methods: sampling, inference and stochastic approximation

Sep 05, 2016 - Sep 08, 2016

ICMS, 15 South College Street

Organisers

Name Institution
Beskos, Alexandros University College London
dos Reis, Gonçalo University of Edinburgh
Gibson, Gavin Heriot-Watt University
Ottobre, Michela Heriot-Watt University
Szpruch, Lukasz University of Edinburgh

The increasing popularity of MCMC algorithms and the need to tackle problems of growing complexity have brought high demands on the efficiency of this class of sampling methods, which are often undeniably costly. The answer to such demands has produced both a higher level of sophistication in the design of MCMC algorithms and the introduction of alternative approaches. In particular, the sampling and stochastic approximation landscape has been noticeably enriched by the introduction of Sequential Monte Carlo (SMC) algorithms and (stochastic) Cubature Methods (CM), both of them most often practically implemented in conjunction with the by now renowned particle methods (and with MCMC as well).

The aim of this workshop is twofold:

  1. to expose the participants and interested faculty to the most recent developments in the field of statistical sampling and stochastic approximation;
  2. to have a truly interdisciplinary event, with the objective of fostering interactions between scientists involved in theoretical developments of sampling methods and those that use such algorithms in applied research.

Photographs of the workshop group are available for download from our Flickr account.

 

 

Arrangements

Participation

All invited speakers should have received their invitation from ICMS. For anyone interested in attending this workshop please complete the public application form on the website.

Venue and Talks

The workshop will be held at 15 South College Street, Edinburgh.  You may also find this map useful for the workshop.  All lectures will be held in the Newhaven Lecture Theatre.    The Lecture Theatre is equipped with a data projector, computer, visualiser (the new generation of overhead projectors) and two blackboards.  The projector and one board may be used simultaneously.  It is best to bring your presentation on a memory stick to use in our ICMS computer.   Alternatively, it is possible for you to use your own laptop with our data projector, but please be aware that you may have to alter your laptop resolutions/settings. 

UK Visas

If you are travelling from overseas you may require an entry visa. A European visa does not guarantee entry to the UK. Please use this link to the UK Visas site to find out if you need a visa and if so how to apply for one.

Travel

Information about travel to the UK and Edinburgh is available here.  Please note that it is your responsibility to have adequate travel insurance to cover medical and other emergencies that may occur on your trip.

A taxi directly from the airport will cost approximately 20.00 GBP to the city centre for a one-way journey. There is also a bus service direct from the airport to the city centre which will cost 4.00 single or 7.00 GBP return - the Airlink 100.  This is a frequent service (every 10 minutes during peak times) and will bring you close to Waverley Railway Station and the workshop venue. 

Lothian buses charge £1.60 for a single, £4.00 for a day ticket. Please note that the exact fare is required and no change is given.

If travelling by train, please note that Edinburgh has several railway stations; Waverley Railway Station being the main station and closest to the workshop venue at 15 South College Street. If you alight at Edinburgh Waverley, the meeting venue is an easy 10 minute walk over North and South Bridge.  The other railway stations are Haymarket and Edinburgh Park but please note that these stations are at the West End of the city centre.

Accommodation 

ICMS will organise accommodation for invited participants.

Catering

The workshop grant will cover refreshments throughout the event and lunch on Monday, Tuesday, Wednesday and Thursday. There will be a workshop dinner on Tuesday evening, and a public lecture and wine reception one evening. 

Wireless Access

Access to wifi via Eduroam is available throughout the building. If you are not registered with Eduroam you will be given instructions and a code for accessing the wireless network.  For those without laptops, there will also be a couple of computers available for you to check your emails.  

Registration Fee

A 70.00 GBP registration fee is payable.

Programme

This is a Provisional Programme and may be subject to alteration.

Abstracts are now available for download.

Monday 5th September 2016

09.00-09.30

Registration & welcome in the Chapterhouse, Level 1, with coffee

09.30-11.00

Arnaud Doucet (University of Oxford)
Minicourse - Part 1

11.00-11.30

Coffee/Tea in the Chapterhouse

11.30-12.15

Claudia Shillings (University of Warwick)
Analysis of the Ensemble Kalman Filter

12.15-14.00

Lunch provided in the Chapterhouse

14.00-14.45

Daniel Clark (Heriot-Watt University)
Modelling and estimating multiple objects

14.45-15.30

Grigorios Pavliotis (Imperial College London)
Optimal Langevin samplers

15.30-16.00

Coffee/Tea in the Chapterhouse

16.00-16.45

Daniela de Angelis (Medical Reserch Council)
Towards computationally efficient inference for influenza dynamic models
pdf of presentation

16.45-17.30

Sebastian Reich (University of Potsdam)
Linear ensemble transform filters for models with non-Gaussian likelihood

17.30-18.30

Informal wine reception in the Chapterhouse


Tuesday 6th September 2016

09.30-10.45

Arnaud Doucet (University of Oxford)
Minicourse - Part 2

10.45-11.15

Coffee/Tea in the Chapterhouse

11.15-12.00

Antonietta Mira (Università della Svizzera Italiana, Switzerland and Università dell'Insubria, Italy)
Adaptive Incremental Mixture Markov chain Monte Carlo

12.00-14.00

Lunch provided in the Chapterhouse

14.00-14.45

Sumeetpal Singh (University of Cambridge)
Blocking strategies and stability of particle Gibbs samplers

14.45-15.30

Konstantinos Spiliopoulos (Boston University)
Irreversible Langevin samplers and variance reduction: a large deviations approach and diffusion on graphs

15.30-16.00

Coffee/Tea in the Chapterhouse

16.00-16.45

Mike Christie (Heriot-Watt University)
Multi-level Hamiltonian Monte Carlo for quantifying uncertainty in reservoir simulation

16.45-17.30

Sebastian Vollmer (University of Oxford)
The bouncy particle sampler: a non-reversible rejection-free Markov chain Monte Carlo method

18.15

Conference dinner at Blonde Restaurant, Edinburgh

 

Wednesday 7th September 2016

09.30-11.00

Dan Crisan (Imperial College London)
Minicourse - Part 1
pdf of presentation

11.00-11.30

Coffee/Tea in the Chapterhouse

11.30-12.15

Prashant Mehta (University of Illinois at Urbana-Champaign)
Gain function approximation in feedback particle filter

12.15-14.00

Lunch provided in the Chapterhouse

14.00-14.45

Joris Bierkens (University of Warwick)
Super-efficient sampling using zig zag Monte Carlo
pdf of presentation

14.45-15.15

Coffee/Tea in the Chapterhouse

15.15-16.00

Gavin Gibson (Heriot-Watt University)
Data augmentation and its role in Bayesian assessment of spatio-temporal models
pdf of presentation

16.00-16.45

Peter Jan Van Leeuwen (University of Reading)
Implicit localisation for particle filters in high-dimensional geophysical systems

17.30-18.00

Doors open to public for public lecture

18.00-19.00

Public Lecture by Gareth Roberts (University of Warwick)
From tea to infinity: a story of randomised algorithms

19.00

Informal wine reception in the Chapterhouse

 

Thursday 8th September 2016

09.00-10.30

Dan Crisan (Imperial College London)
Minicourse - Part 2
pdf of presentation

10.30-11.00

Coffee/Tea in the Chapterhouse

11.00-11.45

Nikolas Kantas (Imperial College London)
Towards particle filtering for signals arising from dissipative stochastic PDEs

11.45-12.30

Darren Wilkinson (Newcastle University)
Scalable algorithms for Markov process parameter inference
pdf of presentation

12.30-14.00

Lunch provided in the Chapterhouse

14.00

Close of workshop

 

Abstracts are now available for download.

 

Presentations:

Presentation Details
Bierkens, Joris
Super-efficient sampling using Zig Zag Monte Carlo
View Abstract Down
Christie, Mike
Multilevel hamiltonian monte carlo for quantifying uncertainty in reservoir simulation
View Abstract Down
Clark, Daniel
Modelling and estimating multiple objects
View Abstract Down
Crisan, Dan
Particle filters in the continuous time framework. Applications of cubature methods
View Abstract Down
De Angelis, Daniela
Towards computationally efficient inference for influenza dynamic models
View Abstract Down
Doucet, Arnaud
Novel pseudo-marginal algorithms for Bayesian inference in latent variable models
View Abstract Down
Gibson, Gavin
Data augmentation and its role in Bayesian assessment of spatio-temporal models
View Abstract Down
Kantas, Nikolas
Towards particle filtering for signals arising from dissipative stochastic PDEs
View Abstract Down
Mehta, Prashant
Gain function approximation in feedback particle filter
View Abstract Down
Mira, Antonietta
Adaptive Incremental Mixture Markov chain Monte Carlo
View Abstract Down
Pavliotis, Grigorios
Optimal Langevin samplers
View Abstract Down
Reich, Sebastian
Linear ensemble transform filters for models with non-Gaussian likelihood
View Abstract Down
Roberts, Gareth
From tea to infinity: a story of randomised algorithms
View Abstract Down
Schillings, Claudia
Analysis of the Ensemble Kalman Filter
View Abstract Down
Singh, Sumeetpal
Blocking strategies and stability of particle gibbs samplers
View Abstract Down
Spiliopoulos, Konstantinos
Irreversible langevin samplers and variance reduction: a large deviations approach and diffusion on graphs
View Abstract Down
van Leeuwen, Peter Jan
Implicit localisation for particle filters in high-dimensional geophysical systems
View Abstract Down
Vollmer, Sebastian
The bouncy particle sampler: a non-reversible rejection-free markov chain monte carlo method
View Abstract Down
Wilkinson, Darren
Scalable algorithms for Markov process parameter inference
View Abstract Down

Participants

Name Institution
Anupam, Angesh University of Sheffield
Beskos, Alexandros University College London
Bierkens, Joris University of Warwick
Christie, Mike Heriot-Watt University
Clark, Daniel Heriot-Watt University
Crisan, Dan Imperial College London
Davidson, Ross SRUC/Bioss
De Angelis, Daniela MRC Biostatistics Unit
dos Reis, Gonçalo University of Edinburgh
Doucet, Arnaud University of Oxford
Duncan, Andrew Imperial College London
Fursov, Ilya Heriot-Watt University
Gibson, Gavin Heriot-Watt University
Jamieson, Magnus University of Strathclyde
Kadochnikova, Anastasia University of Sheffield
Kantas, Nikolas Imperial College London
King, Ruth University of Edinburgh
Lang, Oana-Andrea Imperial College London and University of Reading
Leimkuhler, Ben University of Edinburgh
Lord, Gabriel Heriot Watt University
Maire, Florian University College Dublin
Mehta, Prashant University of Illinois at Urbana-Champaign
Mercieca, Julian University of Sheffield
Mira, Antonietta Università della Svizzera Italiana, Switzerland and Università dell’Insubria, Italy
Ntakaris, Adamantios Tampere University of Technology
Nuesken, Nik Imperial College London
Ottobre, Michela Heriot-Watt University
Pavliotis, Grigorios Imperial College London
Pollock, Murray Warwick
Prentice, Jamie University of Glasgow
Reich, Sebastian University of Potsdam
Roberts, Gareth University of Warwick
Sachs, Matthias University of Edinburgh
Schillings, Claudia University of Warwick
Singh, Sumeetpal University of Cambridge
Spiliopoulos, Konstantinos Boston University
Szpruch, Lukasz University of Edinburgh
Tindemans, Simon Imperial College London
Uney, Murat University of Edinburgh
van Leeuwen, Peter Jan University of Reading
Vollmer, Sebastian University of Oxford
Wilkinson, Darren Newcastle University