The CSML group meets every week to discuss papers of interest. Several times each term a member of the group or an external speaker will present some of their own work or a paper of particular interest; speakers, titles and abstracts are given below.
Computational Statistics and Machine Learning Group
26/1 2pm PSC A54
Chris Sherlock Delayed-acceptance MCMC with examples: advantages and pitfalls and how to avoid the latter [Abstract]
6/12 2pm PSC A54
Jack Baker An overview of Bayesian non-parametrics [Abstract]
17/11 2pm PSC A54
Wentao Li Improved Convergence of Regression Adjusted Approximate Bayesian Computation [Abstract]
20/10 2pm PSC Lab2
Paul Fearnhead The Scalable Langevin Exact Algorithm: Bayesian Inference for Big Data [Abstract]
2/7 2pm B49
Adam Johansen The iterated auxiliary particle filter [Abstract]
19/5 2pm PSC Lab2
Chris Sherlock Pseudo-marginal MCMC using averages of unbiased estimators [Abstract]
9/5 2pm PSC LT
Joris Bierkens (Warwick University) Super-efficient sampling using Zig Zag Monte Carlo [Abstract]
14/4 2pm PSC Lab1
Paul Fearnhead Research opportunities with MCMC and Big Data
17/3 2pm PSC LT
Peter Neal Optimal scaling of the independence sampler [Abstract]
25/2 2pm PSC LT
Paul Fearnhead Continuous-Time Importance Sampling (and MCMC) [Abstract]
18/2 2pm PSC LT
Borja de Balle Pigem Differentially Private Policy Evaluation [Abstract]
10/12 2pm PSC LT
Jack Baker STAN [Abstract]
- 26/11 2pm PSC Lab2
Paul Fearnhead Discussion of "The Bouncy Particle Sampler: A Non-Reversible Rejection-Free Markov Chain Monte Carlo Method" [arXiv] and of "A Markov Jump Process for More Efficient Hamiltonian Monte Carlo" [arXiv].
- 15/10 2pm PSC LT.
James Hensman. Variational inference in Gaussian process models [Abstract]
- 19/05 12pm PSC LT.
Alexandre Thiery (National University of Singapore). Asymptotic Analysis of Random-Walk Metropolis on Ridged Densities [Abstract].
- 28/04 1pm PSC LT.
Chris Sherlock. Delayed acceptance particle marginal random walk Metropolis algorithms
and their optimisation [Abstract].
- 05/03 2pm PSC LT.
Chris Nemeth. Bayesian Inference for Big Data: Current and Future Directions [Abstract].
- 18/12 2pm PSC LT.
Wentao Li. Discussion of the RSS read paper: "Sequential Quasi Monte Carlo" by Mathieu Gerber and Nicolas Chopin.
- 28/11 11am PSC LT.
Chris Nemeth. Particle Metropolis adjusted Langevin algorithms [Abstract].
- 11/03 2pm PSC LT.
Reparameterisations for Particle MCMC [Abstract].
- 25/02 2pm PSC LT.
Vasileios Maroulas (University of Tennessee). Filtering, drift homotopy and target tracking [Abstract].
- 11/12 1pm PSC LT.
Dennis Prangle (Universtity of Bristol). Speeding ABC inference using early-stopping simulations [Abstract].
- 09/05 2pm PSC LT.
Chris Sherlock. Properties and Optimisation of the Pseudo Marginal RWM.
- 17/04 2pm PSC LT.
Anthony Lee (University of Warwick). Particle Markov chain Monte Carlo and marginal likelihood estimation: strategies for improvement.
Dennis Prangle. Likelihood-free parameter estimation for state space models
Joe Mellor (University of Manchester). Thompson Sampling in Switching Environments with Bayesian Online Change Point Detection
21/06 @ 12.00pm (A54 Lecture Theatre)
Classification in Dynamic Streaming Environments - Nicos Pavlidis (Dept of Management Science, Lancaster University).
06/06 @ 12.00pm (A54 Lecture Theatre)
Hamiltonian Monte Carlo: Beyond Kinetic Energy - Paul Fearnhead.
22/05 @ 12.00pm (A54 Lecture Theatre)
Metropolis Adjusted Langevin Algorithm (MALA), simplified Manifold MALA, and Hamiltonian Monte Carlo: motivation, explanation and application - Chris Sherlock.
14/02 @ 12.00pm (A54 Lecture Theatre)
Summary statistics for ABC model choice - Dennis Prangle. [Abstract].
13/12 @ 3.00pm (A54 Lecture Theatre)
Constructing summary statistics for approximate Bayesian computation: semi-automatic ABC
- Paul Fearnhead. [arXiv].
16/11 @ 12.00pm (B35 John Nelder Room)
High-dimensional variable selection via tilting - Haeran Cho (London School of Economics). [Abstract]
17/06 @ 12.00pm (Lab1 PSC)
Online inference and model selection using sequential Monte Carlo - Gareth Ridall. [Abstract].
24/05 @ 12.00pm (A54 Lecture Theatre)
Simulation of mixed speed biochemical reactions using the linear noise
approximation - Chris Sherlock
15/03 @ 12.00pm (Lab1 PSC)
Reading group on "An explicit link between Gaussian fields and
Gaussian Markov random fields:
The SPDE approach" - led by Paul Fearnhead. Preprint.
15/02 @ 12.15pm (Lab1 PSC)
Quantum Monte Carlo - Neil Drummond (Dept of Physics, Lancaster University)
18/01 @ 1.00pm (A54 Lecture Theatre)
Optimal detection of changepoints with a linear computational cost - Rebecca Killick
7/12 @ 12.00pm (A54 Lecture Theatre)
Using ABC for sequential Bayesian analysis - Dennis Prangle.
10/11 @ 12.00pm (A54 Lecture Theatre)
Exact Inference for a Markov switching diffusion
model with discretely observed data - Krzysztof Latuszynski (University of Warwick).
3/11 @ 12.00pm (A54 Lecture Theatre)
Bayesian variable selection using Lasso - Anastasia Lykou.
12/10 @ 12.00pm (A54 Lecture Theatre)
Reading group on "Riemann manifold Langevin and Hamiltonian Monte Carlo methods" - led by Paul Fearnhead. [arXiv].
Filters for models with fixed parameters - Paul Fearnhead
Reading group on Particle MCMC and the pseudo marginal algorithm - led by Gareth Ridall. Giorgos' guide to the pseudo marginal algorithm; Ben Taylor's slides on PIMH; Dennis Prangle's slides on PIMH.
01/12 (lab1 PSC) - The random walk Metropolis: general criteria for the 0.234 acceptance rate rule - Chris Sherlock
03/11 - Likelihood based inference for discretely observed diffusions - Giorgos Sermaidis
20/10 - Sequential Importance Sampling for General Diffusion Models - Paul Fearnhead
28/4 - Reading group on The Integrated Nested Laplace Approximation of Rue et al. (2009) - led by Chris Sherlock. Link to Rue Martino and Chopin (2009); suggested initial reading: Rue and Martino (2007).
02/12 - change point models and fault detection - Paul Fearnhead
28/10 - Optimal scaling of the random walk Metropolis - Part 1 - Chris Sherlock
02/6 - Adaptive Sequential Monte Carlo Methods For Static Inference in Bayesian Mixture Analysis - Ben Taylor
13/5 - The linear least squares prediction view of conjugate gradients - Joe Whittaker
18/3 - Perfect sampling for Random Trees - Hongsheng Dai
04/3 - An MCMC method for Approximate Bayesian Computation - Dennis Prangle
05/2 - Bayesian Analysis of ARMA and Transfer Function Time Series Models - Paul Smith
28/11 - Power sums of lognormals - Chris Sherlock
21/11 - Asymptotic simultaneous bootstrap confidence bounds for simple linear regression lines - Thomas Jaki
31/10 - Using particle filters within MCMC - Paul Fearnhead
Last updated 6/12/2013