Scientific Organisers:
-
Víctor Elvira, University of Edinburgh
-
Jana de Wiljes, University of Potsdam
-
Dan Crisan, Imperial College London
About:
Sequential Monte Carlo (SMC) methods, also known as particle filters or particle methods, have become popular and powerful tools for computational inference in complex probabilistic models used in many and varied fields and applications. The research community includes practitioners and theoreticians at the intersection of statistics, computer science, electrical engineering, and applied mathematics. The interest in SMC methods have rapidly grown in the last decade, jointly with the new challenges and the enormous potential of SMC to tackle high-impact problems in applied sciences (e.g., meteorology, biomedicine, robotics, etc.). The main objectives of the meeting are:
1. to bring together researchers developing and using SMC methods in a diversity of scientific and engineering fields, both in academia and the industry, and
2. to open the field to researchers and users who are new to SMC methods. We will emphasize the interaction of SMC with related areas of research (such as machine learning or data science) where computational inference plays a key role.
Below, we provide a list of key theoretical and methodological topics, as well as application areas.
Theory & Methodology
|
Applications
|
Finally, this event is linked to the Summer School on Bayesian filtering: fundamental theory and numerical methods (SSBF), which will be held also at ICMS on 6-10 May 2024 (i.e., the week before this workshop).
Participation:
SMC 2024 has reached maximum capacity for in-person attendance. All registered participants have been contacted by ICMS with full details including the link to pay the registration fee. If you have not received this email, please check your junk/spam inbox or get in touch.
Due to high demand, this event will now run in a hybrid format. Registration for online attendance will close Friday 3 May 2024, 15.00 BST. You will also receive the Zoom link by 17.00 BST on this day. You may need to check your junk/spam inbox for this email. Please note, online attendance will allow you to watch/interact during talks only. There are no hybrid facilities in the catering area.
Scam Warning:
We are aware of a number of current scams targeting participants at ICMS workshops in relation to registration or accommodation bookings. If you are approached by a third party (eg travellerpoint.org, Conference Committee or Royal Visit) asking for booking or payment details, please ignore.
Please remember:
i) ICMS never uses third parties to do our administration for events: messages will come directly from ICMS staff
ii) ICMS will never ask participants for credit card or bank details
iii) If you have any doubt about an email you receive please get in touch
Programme
MONDAY 13 MAY 2024 | |||
09.00 - 09.25 | Registration and Refreshments | ||
09.25 - 09.30 | Welcome and Housekeeping | ||
09.30 - 10.15 | Christian Robert, Université Paris Dauphine PSL & University of Warwick | Sampling advances by adaptive regenerative processes and importance Monte Carlo | |
10.15 - 11.00 | Sahani Pathiraja, UNSW Sydney | ||
11.00 - 11.30 | Refreshments | ||
11.30 - 12.15 | Joaquín Míguez, Universidad Carlos III de Madrid | ||
12.15 - 13.00 | Axel Finke, Loughborough University | Particle-MALA and Particle-mGRAD: Gradient-based MCMC methods for high-dimensional state-space models | |
13.00 - 14.30 | Lunch | ||
14.30 - 15.15 | Francesca Crucinio, King's College London | A connection between Tempering and Entropic Mirror Descent | |
15.15 - 16.00 | Fredrik Lindsten, Linköping University | ||
16.00 - 16.30 | Refreshments | ||
16.30 - 17.45 | Poster session 1 | ||
TUESDAY 14 MAY 2024 | |||
09.00 - 09.30 | Refreshments | ||
09.30 - 10.15 | Christophe Andrieu, University of Bristol | Monte Carlo sampling with integrator snippets | |
10.15 - 11.00 | Anthony Lee, University of Bristol | Mixing time of the conditional backward sampling particle filter | |
11.00 - 11.30 | Refreshments | ||
11.30 - 12.15 | Arnaud Doucet, University of Oxford | ||
12.15 - 13.00 | Meetings & discussion | ||
13.00 - 14.30 | Lunch | ||
14.30 - 15.20 | Yunpeng Li, University of Surrey | Normalising flow-based differentiable particle filters | |
15.20 - 16.10 | Neil Chada, Heriot Watt University | Multilevel Bayesian Deep Neural Networks | |
16.10 - 16.40 | Refreshments | ||
16.40 - 17.55 | Poster session 2 | ||
19.00 - 22.00 | Conference Dinner | ||
WEDNESDAY 15 MAY 2024 | |||
09.00 - 09.30 | Refreshments | ||
09.30 - 10.15 | Nick Whiteley, University of Bristol | Consistent and fast inference in compartmental models of epidemics using Poisson Approximate Likelihoods | |
10.15 - 11.00 | Jana de Wiljes, TU Ilmenau | ||
11.00 - 11.30 | Refreshments | ||
11.30 - 12.15 | Oana Lang, Imperial College London | ||
12.15 - 13.00 | Meetings & discussion | ||
13.00 - 14.30 | Lunch | ||
14.30 - 15.20 | Nicola Branchini, University of Edinburgh | ||
15.20 - 16.10 | Simo Särkkä, Aalto University | Parallel filtering and smoothing methods for state-space models | |
16.10 - 16.40 | Refreshments | ||
16.40 - 17.55 | Poster session 3 | ||
THURSDAY 16 MAY 2024 | |||
09.00 - 09.30 | Refreshments | ||
09.30 - 10.15 | Pierre Del Moral, Inria | ||
10.15 - 11.00 | Alexandros Beskos, University College London | Antithetic Multilevel Methods for Elliptic and Hypo-Elliptic Diffusions | |
11.00 - 11.30 | Refreshments | ||
11.30 - 12.15 | Daniel Paulin, University of Edinburgh | Unbiased Kinetic Langevin Monte Carlo with Inexact Gradients | |
12.15 - 13.00 | Adam Johansen, University of Warwick | Divide and Conquer Sequential Monte Carlo: Some Properties and Application | |
13.00 - 14.30 | Lunch | ||
14.30 - 16.10 | Meetings & discussion | ||
FRIDAY 17 MAY 2024 | |||
09.00 - 09.30 | Refreshments | ||
09.30 - 10.15 | Sara Pérez Vieites, Aalto University | ||
10.15 - 11.00 | Marcelo Gomes da Silva Bruno, ITA, Brazil | Sequential Monte Carlo Methods for Distributed Bayesian Filtering on Manifolds | |
11.00 - 11.30 | Refreshments | ||
11.30 - 12.15 | Jeremy Heng, ESSEC Business School | Computational Doob's h-transforms for online filtering | |
12.15 - 13.00 | Nicolas Chopin, ENSAE Paris, IPP | Unbiased estimation of smooth functions | |
13.00 - 14.30 | Lunch and end of workshop |