News

Yixuan delivered a talk in the 4th IMA Conference on The Mathematical Challenges of Big Data: GNNs for Node Clustering in Signed and Directed Networks

Two papers are accepted to NeurIPS 2022: A Kernelised Stein Statistic for Assessing Implicit Generative Models and AgraSSt: Approximate Graph Stein Statistics for Interpretable Assessment of Implicit Graph Generators.

Anastasia gave a talk on COMPSTAT in Bologna, Italy: Unsupervised attack pattern detection in cyber-security using topic modelling

Yixuan presented her paper on ICML 2022: GNNRank: Learning Global Rankings from Pairwise Comparisons via Directed Graph Neural Networks

In Stein’s Method: the Golden Anniversary, Anum and Wenkai both gave short talks.

Anastasia presented a poster at ISBA for her paper Bayesian model-based clustering for multiple network data

Tadas gave a talk at ATMCS 2022 titled “Multivariate Normal Approximations for Simplex Counts in Random Complexes”

In Stein’s Method: the Golden Anniversary, Gesine gave a talk on Stein’s method for exponential random graph models and assessing goodness of fit

In Stein’s Method: the Golden Anniversary, Tadas gave a talk on Multivariate Central Limit Theorems for Random Clique Complexes

The paper GNNRank: Learning Global Rankings from Pairwise Comparisons via Directed Graph Neural Networks has been accepted for ICML 2022. Congratulations!

Wenkai will present his paper on AISTATS 2022 Standardisation-function Kernel Stein Discrepancy: A Unifying View on Kernel Stein Discrepancy Tests for Goodness-of-fit

Wenkai will give a tutorial on Introduction to Kernel Methods and a poster presentation for the RSS workshop. The Workshop is organised by Chris Oates and Gesine Reinert.

On Wednesday of even weeks during term time, our group host a reading group on Graph Representation Learning.

Yixuan and Piotr will give talks on The 10th International Conference on Complex Networks and their Applications (CNA).

Stefanos is invited to give a talk on 7th SIGKDD Workshop on Mining and Learning from Time Series (MiLeTS).