Skip to main content

Algorithms & Computationally Intensive Inference seminars

Terms 1-3, Location MB2.23, Fridays 12:15-14:00 (12:15-12:45 is an informal sandwich lunch).

Reminder emails are not sent to participants unless there is a change to the scheduled programme at short notice. If you would like to speak, or you want to be included in any emails, please contact one of the organisers.

Current Organisers: Dootika Vats,

  • If you would like to talk, or have ideas for possible speakers, then please email one of the organisers above.

Website URL:

Mailing List Sign-Up:

Mailing List: (NB - only approved members can post)

2018/19 Term 1:

  • Week 1 - 5th October - Cancelled
  • Week 2 - 12th October - Summary of research short talks:
  • Week 3 - 19th October - Summary of research short talks:
  • Week 4 - 26th October - Giulio Morina (Warwick) - "From the Bernoulli Factory to a Dice Enterprise via Perfect Sampling of Markov Chains"
    • Abstract: Given a p-coin that lands heads with unknown probability p, we wish to construct an algorithm that produces an f(p)-coin for a given function f:(0, 1)→(0, 1). This problem is commonly known as the Bernoulli
      Factory and generic ways to design a practical algorithm for a given function f exist only in a few special cases. We present a constructive way to build an efficient Bernoulli Factory when f(p) is a rational function. Moreover, we extend the original problem to a more general setting where we have access to an m-sided
      die and we wish to roll a v-sided one, where the probability of rolling each face is a fixed function of the original probabilties. We achieve this by perfectly simulating from the stationary distribution of a
      certain class of Markov chains.
  • Week 5 - 2nd November - Cancelled. CoSInES Launch Day.
  • Week 6 - 9th November - Iker Perez (Nottingham) - "Novel approaches to efficiently augment a Markov Jump process for exact Bayesian Inference"
    • Abstract: In this talk we will discuss foundational statistical challenges associated with families of Markovian jump models, which often find applications in domains such as genetics, epidemiology, mathematical biology or queueing theory. We will first review Markov jump processes, and by means of common accessible examples, illustrate the computational impediments posed by real-world application scenarios to inverse uncertainty quantification tasks. Then, we will give an overview of the recent advances linked to structured jump systems. Our work is concerned with building on uniformization procedures and we propose a novel efficient auxiliary-variable algorithm for data augmentation, which yields computationally tractable distributions suited for exact (in Monte Carlo sense) Bayesian inference in often large, infinite or multivariate population systems. We demonstrate the capabilities of the presented methods by drawing Bayesian inference for partially observed stochastic epidemics and show that it overcomes the limitations of existing vanilla approaches. This is joint work (in progress) with Theo Kypraios.
  • Week 7 - 16th November - Ritabrata Dutta (Warwick) - "Well-Tempered Hamiltonian Monte Carlo on Active-Space"
    • Abstract: When the gradient of the log-target distribution is available, Hamiltonian Monte
      Carlo (HMC) has been proved to be an efficient simulation algorithm. However,
      HMC performs poorly when the target is high-dimensional and has multiple isolated
      modes. To alleviate these problems we propose to perform HMC on a locally and
      continuously tempered target distribution. This tempering is based on an efficient
      approach to simulate molecular dynamics in high-dimensional space, known as well-
      tempered meta-dynamics. The tempering we suggest is performed locally and only
      along the directions of the maximum changes in the target which we identify as
      the active space of the target. The active space is the span of the eigenfunctions
      corresponding to the dominant eigenvalues of the expected Hessian matrix of the
      log-target. To capture the state dependent non-linearity of the target, we iteratively
      estimate the active space from the most recent batch of samples obtained from the
      simulation. Finally, we suggest a re-weighting scheme based on path-sampling to
      provide importance weights for the samples drawn from the continuously-tempered
      distribution. We illustrate the performance of this scheme for target distributions
      with complex geometry and multiple modes on high-dimensional spaces in comparison
      with traditional HMC with No-U-Turn-Sampler.
  • Week 8 - 23rd November - Stephen Connor (York) - Title / Abstract TBA
  • Week 9 - 30th November - Emilia Pompe (Oxford) - Title / Abstract TBA
  • Week 10 - 7th December - Flávio Gonçalves (UFMG) - Title / Abstract TBA

2018/19 Term 2:

  • Week 1 - 11th January- Neil Chada (NUS) - Title / Abstract TBA
  • Week 2 - 18th January - James Flegal (UC Riverside) - Title / Abstract TBA (in room MB2.22)
  • Week 3 - 25th January - Joint Session (Short Talks)
  • Week 4 - 1st February - Mateusz Majka (Warwick) - Title / Abstract TBA
  • Week 5 - 8th February - Ioannis Kosmidis (Warwick) - Title / Abstract TBA
  • Week 6 - 15th February - Available 
  • Week 7 - 22nd February - Patrick Rebeschini (Oxford) - Title / Abstract TBA
  • Week 8 - 1st March - Susana Gomes (Warwick) - Title / Abstract TBA
  • Week 9 - 8th March - Available 
  • Week 10 - 15th March - Available 

2018/19 Term 3:

  • Week 1 - 26th April - Available
  • Week 2 - 3rd May - Available 
  • Week 3 - 10th May - Available 
  • Week 4 - 17th May - Available 
  • Week 5 - 24th May - Available 
  • Week 6 - 31st May - Available 
  • Week 7 - 7th June - Available 
  • Week 8 - 14th June - Available 
  • Week 9 - 21st June - Available 
  • Week 10 - 28th June - Available 

Previous Years:









Some key phrases:

- Sampling and inference for diffusions
- Exact algorithms
- Intractable likelihood
- Pseudo-marginal algorithms
- Particle filters
- Importance sampling
- Adaptive MCMC

- Perfect simulation

- Markov chains...
- Random structures...
- Randomised algorithms...