• Day 1 Monday, May 20
    8:00- 9:00 Arrivals

    9:00- 9:10 Welcome Magnus Rattray, University of Manchester Data Science Institute

    9:10- 9:45 Language, meaning and meta-embeddings Danushka Bollegala, University of Liverpool

    9:50- 10:25 On the use and interpretation of neural attention mechanisms for biomedical natural language processing Byron Wallace, Northeastern University, USA

    10:30 Coffee

    11:00- 11:15 Text mining student discussion forum data: common pitfalls and how to avoid them Elaine Farrow, University of Edinburgh

    11:20- 11.35 Towards an Emoji WordNet Matthew Shardlow, Manchester Metropolitan University

    11.40- 12:15 Coping with large-scale terminology in a virtual patient dialogue system Pierre Zweigenbaum, CNRS, Paris

    12:20 Lunch and posters

    13:30- 14:05 Bespoke machine learning for humanoid robots Georgio Metta, IIT Italian Institute of Technology, Genoa, Italy

    14:10- 14:45 Statistical hypothesis testing in a Bayesian non-parametric framework Sarah Filippi, Imperial College London

    14:50- 15:05 Predicting progression in heterogeneous neurodegenerative diseases using a joint mixture model approach Frank Dondelinger, Lancaster University

    15:10 Coffee

    15:40- 16:15 Statistical challenges in the analysis of single-cell gene expression data Catalina Vallejos, University of Edinburgh and Alan Turing Institute

    16.20- 16:35 Flexible, wide-coverage normalisation of medical concept mentions Paul Thompson, University of Manchester

    16.40- 16:55 Converting tweets into sequential data to detect depressed users Akkapon Wongkoblap, Kings College London

    17:00 Posters and drinks

    19:30 Dinner at Browns
  • Day 2 Tuesday, May 21
    8:00- 9:00 Arrivals

    9:00- 9:35 Will Crime Science eventually become Data Science? Bennett Kleinberg, University College London

    9:40- 10:15 Old laws for new (algorithmic) tricks Marion Oswald, University of Winchester

    10:20 Coffee

    11:00- 11.35 The Synergy of Behavioural and Data Science in Security Paul Taylor, Lancaster University

    11:40- 11:55 When the Magic Wears Off: Flaws in ML for Security Evaluations (and What to Do about It) Lorenzo Cavallaro, Kings College London

    12:00- 12:15 Adversarial Vulnerability Bounds for Gaussian Process Classification Michael Smith, University of Sheffield

    12:20 Lunch

    13:30- 14:05 Fair Machine Learning Silvia Chiappa, DeepMind

    14:10- 14:45 Artificial Intelligence for Data Analytics Chris Williams, University of Edinburgh and Alan Turing Institute

    14:50- 15:05 Deconfounded Random Forests Reham Badawy, Aston University

    15:10 Coffee

    15:40- 15:55 Decomposing feature-level variation with Covariate Gaussian Process Latent Variable Models Kaspar Märtens, University of Oxford

    16:00- 16:15 ANNABELL integrated in iCub: towards a grounded multi language system Giovanni Masala, Manchester Metropolitan University

    16:20- 16.55 Machine Learning for personalisation in assistive robotics Yiannis Demiris, Imperial College London

    17.00 Close