PASCAL2
PASCAL 2008 Workshop on
Approximate Inference in Stochastic Processes and Dynamical Systems

May 27-29, Cumberland Lodge, UK

[Overview - Speakers - Important dates - Organisers - Programme - Sponsors]

Overview

The modelling of continuous-time stochastic processes from uncertain (discrete) observations is an important task that arises in a wide range of applications, such as in climate modelling, tracking, finance and systems biology. Although observations are in general only available at discrete times, the underlying system is often a continuous-time one. Hence, the physics or the dynamics are formulated by systems of differential equations, the observation noise and the process uncertainty being modelled by several stochastic sources.

When dealing with stochastic processes, it is natural to take a probabilistic approach. For example, we may incorporate prior knowledge about the dynamics by providing probability distributions on the unknown functions. In contrast to models that are only data driven, it is hoped that incorporating domain knowledge in the inference process will improve performance in practice. The main challenges in this context are how to deal with continuous-time objects, how to do inference and how to be agnostic about the deterministic driving forces and the sources of uncertainty.

The workshop will provide a forum for discussing the open problems arising in dynamical systems, and in particular continuous-time stochastic processes. It will focus both on the mathematical aspects/theoretical advances and the applications. Another important aim is to bridge the gap between the different communities (data assimilation, machine learning, optimal control, systems biology, finance, ...) and favour interactions. Hence, the workshop will be of interest to researchers from statistics, computer science, mathematics, physics and engineering. We also hope that the workshop will provide new insights in this exciting field and serve as a starting point for new research perspectives and future collaborations.

We welcome abstract contributions which will be peer-reviewed and selected for presentation. The abstract should not be longer than two A4 pages and in PDF format. It should be sent to ais@cs.ucl.ac.uk by 20 April 2008.

After the workshop, all presentations will be made available on the web as video lectures. Authors will be invited to submit a full paper version of their work for a collective volume that will summarise the major contributions to this meeting.

The workshop is sponsored by PASCAL2 network of excellence and is one of six workshops in the Thematic Programme in Leveraging Complex Prior Knowledge for Learning.

Confirmed speakers
Important dates

Abstract submission deadline: 20 April 2008.
Notification to authors: 30 April 2008.
Registration deadline: 14 May 2008.
Workshop: 27-29 May 2008.

Organisers
Programme

The workshop will take place in Cumberland Lodge, near Windsor (UK). For directions please click here. The workshop will start at 12:30 on Tuesday 27 May and will run until 16:00 on Thursday 29 May. There will be plenty of time for informal discussions.

The talks are now available as videolectures from http://videolectures.net/aispds08_cumberland_lodge/.

Tuesday 27/05
 
12:30 - 14:00 Arrival and lunch
14:00 - 15:00
 
An introduction to Levy processes with financial modelling in mind [abstract] [talk]
Matthias Winkel, University of Oxford
15:00 - 15:50
 
Variational filtering in generated coordinates of motion [abstract] [talk]
Karl Friston, University College London
15:50 - 16:20 Coffee break
16:20 - 16:50
 
Density estimation of initial conditions for populations of dynamical systems [abstract] [talk]
Alberto Busetto, ETH Zurich
16:50 - 17:20
 
Sparse Multi-output Gaussian Processes [abstract] [talk]
Mauricio Alvarez, University of Manchester
17:20 - 18:10
 
Estimating the probability of rare climate events: inference from a large deterministic computer code [abstract] [talk]
Peter Challenor, University of Southampton
18:30 - 21:30
 
Reception and dinner
 
Wednesday 28/05
 
09:30 - 10:20
 
Approximate inference for continuous time Markov processes [abstract] [talk]
Manfred Opper, Technical University Berlin
10:20 - 10:50
 
Variational inference and learning for continuous-time nonlinear state-space models [abstract] [talk]
Tapani Raiko, Helsinki University of Technology
10:50 - 11:20 Coffee break
11:20 - 11:50
 
An efficient Monte-Carlo algorithm for the ML-Type II parameter estimation of nonlinear diffusions [abstract] [talk]
Yuan Shen, Aston University
11:50 - 12:40
 
MCMC schemes for partially observed diffusions - Some recent advances [abstract] [talk]
Andrew Golighlty, Newcastle University
12:45 - 14:00 Lunch
14:00 - 14:50
 
Normalized kernel-weighted random measures [abstract] [talk]
Jim Griffin, University of Kent
14:50 - 15:20
 
Solving the data association problem in multi-object tracking by Fourier analysis on the symmetric group [abstract] [talk]
Risi Kondor, University College London
15:20 - 15:50
 
Approximate Bayesian computation: a simulation based approach to inference [abstract] [talk]
Richard Wilkinson, University of Sheffield
15:50 - 16:20 Coffee break
16:20 - 17:10
 
Exact simulation of jump diffusions [abstract] [talk]
Flavio Goncalves, University of Warwick
17:30 - 18:30 Afternoon walk
19:15 - 21:30
 
Dinner
 
Thursday 29/05
 
09:30 - 10:20
 
An efficient approach to stochastic optimal control [abstract] [talk]
Bert Kappen, Radboud University Nijmegen
10:20 - 10:50
 
Information evolution of optimal learning [abstract] [talk]
Roman Belavkin, Middlesex Univeristy
10:50 - 11:20 Coffee break
11:20 - 11:50
 
Approximate system identification: misfit versus latency [abstract] [talk]
Ivan Markovsky, University of Southampton
11:50 - 12:40
 
Sigma point and particle approximations of stochastic differential equations in optimal filtering [abstract] [talk]
Simo Särkkä, Helsinki University of Technology/Nalco Company
12:45 - 14:00 Lunch
14:00 - 14:30
 
State estimation and prediction based on dynamic spike train decoding: noise, adaptation, and multisensory integration [abstract] [talk]
Ron Meir, Technion
14:30 - 15:00
 
Gaussian process toolkit for modelling the dynamics of transcriptional regulation [abstract] [talk]
Pei Gao, University of Manchester
15:00 - 15:50
 
 
On stratified path sampling of the Thermodynamic Integral: computing Bayes factors for nonlinear dynamical systems models
of biochemical pathways [abstract] [talk]
Mark Girolami, University of Glasgow
15:50 - 16:00
 
End of workshop
 
Sponsors
PASCAL Network