Uncategorised

Prof. Mimi Zhang and I recently had a paper accepted to the International Conference on Data Mining 2021 (ERA rank A, Qualis Rank A1). Our paper introduces Density-Core Finding (DCF), an efficient and robust density-based clustering method. DCF extends the popular Density Peaks Clustering (DPC) method which detects modes as points with high density and … Read more

On December 12th, Prof. Alessio Benavoli presented new work on learning choice functions using a Bayesian model. He included an experiment in the talk, demonstrating an application to multi-objective preferential Bayesian optimisation. Abstract: In practical decision making, there are usually several preference relations that have to be taken into account.  For instance, when choosing among … Read more

On October 21st, Daniel Dempsey presented a paper by Joesph, Du Berger & Bélise entitled ‘Bayesian and Mixed Bayesian/Likelihood Criteria for Sample Size Determination’ to the reading group. Slides summarising the paper are available below.

On October 5th, the seminar series for the Michaelmas term commenced with a talk from Prof. Emilio Porcu of Khalifa University, Abu Dhabi. Prof. Porcu gave an overview of recent developments in spatial statistics, how climate datasets have affected the field, and how reformulating problems in more abstracts terms can make solutions easier to find. … Read more

Athanasios Georgiadis presents two tutorials to the seminar group, introducing recent work in spatial statistics on May 5th, and nonparametric density estimation on May 12th. The abstracts for both sessions are available below. Seminar 1:Recent progress and open directions in spatial statistics  Abstract: Data’s location plays a decisive role in the extraction of statistical decisions. The … Read more

The seminar series continued on April 28th. PhD Candidates Sebastian Appleby and Tobi Olajubu presented introduced their work to the group. Sebastian is studying Sardex, a community currency network launched in 2010 and is aiming at modelling the network growth. Tobi is developing a method for anomaly detection for data of mixed features. Slides from … Read more

On April 7th, the seminar group welcomed Dr. Florian Maire from the Université de Montréal. Florian introduced recent work detailing a method for comparing two or more competing MCMC algorithms. Details for Florian’s talk are below. Title: Weak Peskun ordering for approximate MCMC comparison Abstract: Despite the popularity of MCMC methods in Bayesian statistics and … Read more

On Wednesday, March 24th, James Ng presented recent work related to nonparametric statistical models based on Gaussian processes on the unit hypersphere. An abstract for James’ talk is below. Abstract We study nonparametric statistical models based on Gaussian processes on the unit hypersphere. We derive convergence rates of posterior distributions for various choices of covariance kernels. We then … Read more

On March 10th, the seminar series welcomed Alexandra Gessner, a PhD student at the Max-Planck Institute in Tübingen, Germany. Alexandra introduced her work which presents a novel sampling method that enables rejection-free sampling from linearly constrained Gaussian domains. An abstract for Alexandra’s talk is given below. Abstract Gaussian densities are ubiquitous in statistics, yet Gaussian probabilities … Read more

On March 3rd, the seminar series hosted the second PhD Lightning Round. Laura introduced her work with ‘Diversity Interaction Models’, particularly in regards to multifunctionality and the interpretation of transformations on the response. Joshua introduced a new project investigating methods for Gaussian mixtures in the presence of numerous clusters.