Past events

The second UCL Workshop on the Theory of Big Data took place January 6-8 2016. Archived information about this event is below.

Details of the inaugural 2015 event are further down the page.

Return to the homepage for details of the 2017 event.

Day 1: Wednesday 6th January 2016

09.15 - Registration desk opens
10.00 – 10.05 Welcome from Conference Chair (Prof Sofia Olhede, UCL, UK)
10.05 – 10.15 Big Data Perspectives & The Alan Turing Institute (Prof Andrew Blake, The Alan Turing Institute, UK)
10.15 – 11.15 Plenary Talk, Model based variable clustering: minimax membership detection (Prof Florentina Bunea, Cornell, USA) [Slides]
11.15 – 11.30 Break
11.30 – 12.00 Risks of assuming data equals ‘all’ (Prof David Hand, Imperial College London, UK) [Slides]
12.00 – 12.30 Large Observational Healthcare Studies: Challenges, Pitfalls and Future Directions (Prof David Madigan, Columbia, USA) [Slides]
12.30 – 13.00 Kernel methods for topological data analysis (Prof Kenji Fukumizu, ISM Japan) [Slides]
13.00 – 14.00 Lunch
14.00 – 14.30 Removing Red Herrings and avoiding Wild Goose Chases – Creating Impact (Prof Peter Grindrod, Oxford, UK)
14.30 – 15.00 From Graph Theory to some algorithmic applications (Prof Ken-ichi Kawarabayashi, NII, Japan)
15.00 – 15.30 Contributed Oral Presentations
15.00: Scalability Issues and the Potential for Encrypted Machine Learning (Dr Louis Aslett et al., Oxford, UK) – Abstract
15.15: Random Intersection Trees for finding interactions in large datasets (Dr Rajen Shah et al., Cambridge, UK) – Abstract
15.30 – 15.45 Break
15.45 – 16.15 Inference in high-dimensional nonparametric graphical models (Dr Mladen Kolar, Chicago Booth, USA) [Slides]
16.15 – 16.45 Graphical Model for Time Series (Dr Michael Eichler, Maastricht University, Netherlands) [Slides]
16.45 - 17.15 Big Data Visual Analytics: A User-Centric Approach (Dr Remco Chang, Tufts University, USA)

Day 2: Thursday 7th January 2016

09.00 - Registration desk opens
09.30 – 10.30 Plenary talk: Blessing of heterogeneous large-scale data for high-dimensional causal inference (Prof Peter Bühlmann, ETH, Switzerland) [Slides]
10.30 – 11.00 Structural Optimization: New perspectives for increasing efficiency of numerical schemes (Prof Yurii Nesterov, UCL, Belgium) [Slides]
11.00 – 11.15 Break
11.15 – 11.45 Boosting the rates by adding structural model constraints : application to modeling the electric consumption. (Prof Dominique Picard, Paris VII, France) [Slides]
11.45 – 12.15 Adaptive Estimation of the Copula Correlation Matrix for Semiparametric Elliptical Copulas (Prof Marten Wegkamp, Cornell, USA) [Slides]
12.15 – 12.45 Hierarchical modelling of automated imaging data (Prof Darren Wilkinson, Newcastle) [Slides]
Turing Gateway to Mathematics event: Big Data Analytics for Financial Services
13.00 - 13.45 Registration, Lunch and Tea/Coffee
13.45 - 14.00 Welcome and Introduction. – Jane Leeks (Turing Gateway to Mathematics) / Prof Patrick Wolfe (University College London)
14.00 – 14.30 Big Data, Algorithmic Finance and The future of Financial Modelling (Prof Rama Cont, Imperial College London)
14.30 – 14.55 Data Analytics and Industry Needs (Dickie Whittaker, Oasis Loss Modelling Framework / Lighthill Risk Network)
14.55 – 15.20 High Frequency Trading Behaviours: Data Challenges (Tanya Reeves, Bank of Tokyo Mitsubishi)
15.20 – 15.35 Tea and Coffee
15.35 – 15.55 Data-Driven Financial Conduct Regulation (Stefan Hunt, Financial Conduct Authority)
15.55 – 16.20 Alternative Finance – Who, What, Why, Where and are we Nearly There Yet? (Louise Beaumont, GLI Finance)
16.20 – 16.45 Financial Regulator Perspective (Paul Robinson, Bank of England)
16.45 – 17.15 Questions and Chaired Open Discussion
17.15 – 18.00 Networking and Drinks Reception

Day 3: Friday 8th January 2016

09.00 - Registration desk opens
09.30 – 10.15 Fisher Speaker: Scalable inference for a full multivariate stochastic volatility model (Prof Petros Dellaportas, UCL, UK) [Slides]
10.15 – 10.45 Fast ‘tail-greedy’ bottom-up decomposition algorithms for signals on graphs and network adjacency matrices (Prof Piotr Fryzlewicz, LSE, UK)
10.45 – 11.00 Break
11.00 – 11.30 Functional regression analysis for big motion data (Dr Jian Qing Shi, Newcastle University, UK) [Slides]
11.30 – 12.00 Big Data in Time: Progress and Challenges from Oceanography (Dr Jonathan Lilly, NorthWest Research Associates, USA) [Slides]
12.00 – 12.30 A Timing Approach to Causal Network Inference (Prof Negar Kiyavash, University of Illinois at Urbana–Champaign)
12.30 – 14.30 Lunch and Poster Presentations
14.30 - 15.00 Contributed Oral Presentations
14.30: Computational limits for distributed estimation (Dr Quentin Berthet, Cambridge, UK) – Abstract
14.45: O(N logN) bias-corrected maximum likelihood estimation of stochastic processes with applications to a global oceanographic dataset (Dr Adam Sykulski (UCL), et al.) – Abstract
15.00 Closing Remarks (Prof Sofia Olhede, UCL, UK)

Information on the posters presented at this workshop is also available.


The inaugural Theory of Big Data conference took place in January 2015. Archived information can be found on this page.

Day 1: 7th January 2015

10.00 – 10.10 Introduction
10.10 – 11.10 Maximin effects in inhomogeneous large-scale data. – Prof. Nicolai Meinshausen Slides
11.10 – 11.40 Break
11.40 – 12.20 Random walk models of graph formation. – Dr. Peter Orbanz Slides
12.20 – 13.10 Understanding the Behavior of Large Networks. – Prof. Patrick Wolfe

 

TGM event: Coping with Big Data – an Analytics and Computational Perspective
14.00 – 14.30 Registration and Tea/Coffee
14.30 – 14.35 Welcome and Introduction. – Turing Gateway to Mathematics / University College London Programme
14.35 – 15.05 Statistical Challenges of Big Data – Dr. Niall Adams Slides/Video
15.05 – 15.35 Big Data Challenges for Business. – Giles Pavey Slides/Video
15.35 – 16.05 Computational Limitations and Cloud Area. – Derek McAuley Slides/Video
16.05 – 16.20 Tea and Coffee
16.20 – 16.50 Practical Challenges of Data Analytics. – Leigh Lapworth Slides/Video
16.50 – 17.10 Questions and Chaired Open Discussion Video
17.10 – 18.00 Tea/Coffe and Networking

Day 2 : 8th January 2015

09.00 09.50 Network comparison. – Prof. Gesine Reinert Slides
09.50 – 10.40 Min-wise hashing for large-scale regression. – Dr. Rajen Shah Slides
10.40 – 11.10 Break
11.10 – 12.00 Spectral Methods for Unsupervised and Discriminative Learning with Latent Variables. – Prof. Animashree Anandkumar Slides
12.00 – 12.50 Multistage Bandits. – Dr. Philippe Rigollet
12.50 – 14.00 Lunch
14.00 – 15.10 Oral Presentations:
Estimating Latent Variable Densities For Exchangeable Network Models. – Sharmodeep Bhattacharyya Slides
Distributed Monte Carlo testing. – Patrick Rubin-Delanchy Slides
Nonparametric Eigenvalue-Regularized Precision or Covariance Matrix Estimator. – Clifford Lam Slides
Consistent Vector-valued Distribution Regression. – Zoltan Szabo Slides
15.10 – 17.00 Posters and refreshments

Day 3 : 9th January 2015

09.00 – 09.50 High-dimensional Ordinary Least-squares Projector for Screening Variables. – Prof. Chenlei Leng Slides
09.50 – 10.40 Randomized dual coordinate ascent with arbitrary sampling. – Dr. Peter Richtárik Slides
10.40 – 11.10 Tea
11.10 – 12.00 General Bayesian updating for complex applications – Prof. Chris Holmes Slides
12.00 – 12.50 Large Segmenting Multiple Time Series by Contemporaneous Linear Transformation: PCA for Time Series. – Prof. Qiwei Yao Slides
12.50 – 14.00 Lunch
14.00 – 14.45 Oral Presentations
Big Hypothesis Testing with Kernels. – Dino Sejdinovic Slides
Unbiased Posterior Expectations for Big Data. – Heiko Strathmann Slides
Statistical calculations at scale using machine learning algorithms and emulation. – Daniel John Lawson Slides