MCMC and Particle Methods: Sampling, Inference and Stochastic Approximation

Home > Workshops > 2016 > MCMC and Particle Methods: Sampling, Inference and Stochastic Approximation

MCMC and Particle Methods: Sampling, Inference and Stochastic Approximation

 05 - 08 Sep 2016

ICMS, 15 South College Street Edinburgh

Scientific Organiser

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

About:

he 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 was to expose the participants and interested faculty to the most recent developments in the field of statistical sampling and stochastic approximation; and 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.

Programme:

  • Daniela de Angelis, Medical Reserch Council - Towards Computationally Efficient Inference for Influenza Dynamic Models

  • Antonietta Mira, Università della Svizzera Italiana and Università dell'Insubria - Adaptive Incremental Mixture Markov Chain Monte Carlo

  • Nikolas Kantas, Imperial College London - Towards Particle Filtering for Signals Arising from Dissipative Stochastic PDEs

  • Peter Jan Van Leeuwen, University of Reading - Implicit Localisation for Particle Filters in High-Dimensional Geophysical Systems

  • Mike Christie, Heriot-Watt University - Multi-Level Hamiltonian Monte Carlo for Quantifying Uncertainty in Reservoir Simulation

  • Gavin Gibson, Heriot-Watt University - Data Augmentation and its Role in Bayesian Assessment of Spatio-Temporal Models

  • Sumeetpal Singh, University of Cambridge - Blocking Strategies and Stability of Particle Gibbs Samplers

  • Prashant Mehta, University of Illinois - Gain Function Approximation in Feedback Particle Filter

  • Grigorios Pavliotis, Imperial College London - Optimal Langevin Samplers

  • Claudia Shillings, University of Warwick - Analysis of the Ensemble Kalman Filter

  • Daniel Clark, Heriot-Watt University - Modelling and Estimating Multiple Objects

  • Sebastian Reich , University of Potsdam - Linear Ensemble Transform Filters for Models

  • Joris Bierkens, University of Warwick - Super-Efficient Sampling Using Zig Zag Monte Carlo

  • Darren Wilkinson, Newcastle University - Scalable Algorithms for Markov Process Parameter Inference