Turing Fellows

The Alan Turing Institute and its five partner universities have announced a list of inaugural Turing Fellows.

Below is the full list with clickable links that will direct you to each Fellow’s university home page.

Faculty directors



Cambridge

Professor Jon Crowcroft

Cambridge

Bio

Jon Crowcroft has been the Marconi Professor of Communications Systems in the Computer Laboratory since October 2001. He has worked in the area of Internet support for multimedia communications for over 30 years. Three main topics of interest have been scalable multicast routing, practical approaches to traffic management, and the design of deployable end-to-end protocols. Current active research areas are Opportunistic Communications, Social Networks, and techniques and algorithms to scale infrastructure-free mobile systems. He leans towards a “build and learn” paradigm for research.

He graduated in Physics from Trinity College, University of Cambridge in 1979, gained an MSc in Computing in 1981 and PhD in 1993, both from UCL. He is a Fellow the Royal Society, a Fellow of the ACM, a Fellow of the British Computer Society, a Fellow of the IET and the Royal Academy of Engineering and a Fellow of the IEEE.

He likes teaching, and has published a few books based on learning materials.

Research

Computing Systems at scale are the basis for much of the excitement over Data Science, but there are many challenges to continue to address ever larger amounts of data, but also to provide tools and techniques implemented in robust software, that are usable by statisticians and machine learning experts without themselves having to become experts in cloud computing. This vision of distributed computing only really works for “embarrassingly parallel” scenarios. The challenge for the research community is to build systems to support more complex models and algorithms that do not so easily partition into independent chunks; and to give answers in near real-time on a given size data centre efficiently.

Users want to integrate different tools (for example, R on Spark); don’t want to have to program for fault tolerance, yet as their tasks & data grow, will have to manage this; meanwhile, data science workloads don’t resemble traditional computer science batch or single-user interactive models. These systems put novel requirements on data centre networking operating systems, storage systems, databases, and programming languages and runtimes. As a communications systems researcher for 30 years, I am also interested in specific areas that involve networks, whether as technologies (the Internet, Transportation etc), or as observed phenomena (Social Media), or in abstract (graphs).

Dr Leonardo Bottolo

Cambridge

Bio

Dr Leonardo Bottolo is Reader in Statistics for Biomedicine at the University of Cambridge. He received his PhD in Methodological Statistics from the University of Trento, Italy, in 2001. Before joining the University of Cambridge, he was appointed Senior Lecturer in Statistics in the Department of Mathematics, Imperial College. He worked as postdoc in the Mathematical Genetics group, University of Oxford and at the Institute of Mathematical Sciences, Imperial College.

Research

Dr Leonardo Bottolo has considerable experience of methodological and algorithmic aspects of data integration and feature selection in genetics and system biology for high-dimensional data. In continuity with his experience in population genetics gained in Oxford he implemented new multivariate approaches for the detection of systemic associations between different types of -omics data. Currently he is further developing these statistical tools with novel MCMC samplers to make them fully operational on current large genetics data and on future high-throughput Next Generation Sequencing. The implementation of these algorithms benefits from recent advances in parallel computing of massive data sets.

Dr Quentin Berthet

Cambridge

BIO

Quentin Berthet is a Lecturer in the Statslab, in the DPMMS at Cambridge, and a fellow of St John’s College, since 2015. He is a former student of the Ecole Polytechnique, received a Ph.D. from Princeton University in 2014, and was a CMI postdoctoral fellow at Caltech.

RESEARCH

Dr Berthet’s research is focused on the theoretical understanding of statistical procedures. He studies fundamental problems related to modern datasets such as computational constraints, heterogeneity of the data or presence of errors. He is interested in taking into account societal concerns in the analysis of data, by considering ethical aspects (privacy, security of data), as well as ways to guide decision making in a data-driven way. He uses tools from Statistics and Probability Theory, as well as from Optimisation, Information Theory and Computer Science.

Professor Carl Rasmussen

Cambridge

BIO

Professor Carl Rasmussen is a lecturer in the Machine Learning Group of the Computational and Biological Learning Lab in the Division of Information Engineering at the Department of Engineering in Cambridge.

 

RESEARCH

Carl has very broad interests in probabilistic inference in machine learning, covering both unsupervised, supervised and reinforcement learning. He is particularly interested in design and evaluation of non-parametric methods such as Gaussian processes and Dirichlet processes. Exact inference in these models is often intractable, so we need to resort to approximation methods, such as variational techniques or Markov chain Monte Carlo.

 

 

Professor Richard Samworth

Cambridge

Bio

Richard Samworth is Professor of Statistics in the Statistical Laboratory at the University of Cambridge and a Fellow of St John’s College.  He received his PhD, also from the University of Cambridge, in 2004, and currently holds an EPSRC Early Career Fellowship.

Research

His main research interests are in developing methodology and theory for high-dimensional and nonparametric statistical inference.  He is currently particularly interested in techniques for handling statistical challenges in Big Data that rely on perturbations of the data and aggregation.  Examples include random projection ensembles for high-dimensional classification and subsampling for variable selection. Other interests include shape-constrained and other nonparametric function estimation problems, Independent Component Analysis, and the bootstrap and its variants (e.g. bagging).

Dr Rajen Dinesh Shah

Cambridge

Bio

Dr Rajen Dinesh Shah is a Lecturer in Statistics at the Statistical Laboratory, University of Cambridge, where he previously completed my PhD at Trinity College under the supervision of Professor Richard Samworth. He is a member of the Research Section committee of the Royal Statistical Society.

Research

Dr Shah is broadly interested in the analysis of large-scale data and my work falls under the areas of high-dimensional statistics, machine learning and computational statistics. He primarily works on developing algorithms for learning from large-scale data, and understanding their theoretical properties. He is particularly interested in creating methods that are able to perform well under constraints on computation and memory. Recently he has worked on variable selection and uncertainty quantification in high-dimensional settings; hashing methods for compressing large-scale data; and randomised algorithms for detecting interactions.

Dr Sumeetpal Singh

Cambridge

BIO

Sumeetpal S. Singh received the B.E. (with first-class honours) and Ph.D. degrees from the Dept. of Electrical Engineering, University of Melbourne, Australia, in

1997 and 2002, respectively. After having worked in Industry for a number of years, he joined the Cambridge University Engineering Department in 2004 as a Research Associate and is currently a University Senior Lecturer in Engineering Statistics. He is also a Fellow and Director of Studies at Churchill College, an Affiliated Lecturer of the Statistics Laboratory and an Associate Editor of Statistics and Computing. His research focus is on statistical signal processing, in particular, using Monte Carlo methods, covering algorithmic development for applications and theoretical analysis.  He has been recognised for his work on Multi-target Tracking (awarded the IEEE M. Barry Carlton Award in 2013.)

 

RESEARCH

Broadly, his research develops core statistical methodology for Data Science with the view of using cloud computing techniques. Analysing heterogeneous data sets [which is necessary for data-driven decision making] is a major hurdle in Data Science [e.g. Ecology, Epidemiology] as the mathematical models that describe the data are increasingly more intricate and very high dimensional in most cases. Fortunately, statistical inference is still feasible with Monte Carlo methods. His research is inherently cross-disciplinary as he draw on tools from Statistics, Engineering, Computer Science and Probability Theory to design and analyse new computational procedures.

Dr Richard Gibbens

Cambridge

Bio

Richard Gibbens is a mathematician and computer scientist with research interests in the mathematical modelling of networks especially communication, road transport and energy networks. He is a Reader at the Computer Laboratory, University of Cambridge. Prior to 2001 he was a Royal Society University Research Fellow in the Statistical Laboratory, University of Cambridge where he also obtained his PhD.

Research

During the Alan Turing Institute Faculty Fellowship he will be exploring the interface between mathematics, statistics and computer science. This will involve theoretical modelling reinforced with working with data drawn from, for example, road transport networks and energy networks. He is interested in working on the boundaries of current methodology and discovering how recent advances in computer technology such as multi-core and cluster computing can be best utilised. He is also interested in collaborating widely with researchers and policy makers to ensure that data and modelling are both used and valued.

Professor Peter Markowich

Cambridge

BIO

Professor of Applied Mathematics, Department of Applied Mathematics and Theoretical Physics, University of Cambridge and Professor of Applied Analysis, Faculty of Mathematics, University of Vienna.

 

RESEARCH

Professor Peter Markowich’s research is in Applied Mathematics, in particular in the partial differential equations (PDE) which arise in applications. Mine is an ‘integrated’ approach, involving mathematical
analysis, numerical analysis, computational mathematics and mathematical modeling. He has
spent a significant part of my career working on PDE in physics but about ten years ago he switched his research emphasis to PDE applications in the life, social and data sciences, emphasizing both direct and inverse (data assimilation) problems. This change of research direction shall support and help to shape the current trend in modern mathematics of penetrating and becoming indispensable tools also for non-physical data-rich applied sciences. He has been working on recently: inpainting in mathematical imaging, price formation of economical assets, human crowd motion modeling by fluid and mean-field game approaches, large data assimilation in the Navier-Stokes fluid equations, and network formation and adaptation in the biological and social sciences.

Dr Adrian Weller

Cambridge

BIO

Adrian Weller is a senior researcher in the Machine Learning Group at the University of Cambridge, advised by Prof Zoubin Ghahramani. He received a PhD in computer science (machine learning) at Columbia University, advised by Prof Tony Jebara. He is an active angel investor and advisor, and previously held senior roles in investing and trading at Goldman Sachs, Salomon Brothers and Citadel.

 

RESEARCH

Most of his academic research relates to graphical models but he is also very interested in other areas including: finance, anything on intelligence (natural or artificial), deep learning, reinforcement learning, evolution, Bayesian methods, time series analysis, ethics, music and methods for big data.

Professor Carola-Bibiane Schönlieb

Cambridge

Bio

Carola-Bibiane Schönlieb is a Reader in Applied and Computational Analysis at the Department of Applied Mathematics and Theoretical Physics (DAMTP), University of Cambridge since 2015. There, she is head of the Cambridge Image Analysis group, Director of the Cantab Capital Institute for Mathematics of Information, Co-Director of the EPSRC Centre for Mathematical and Statistical Analysis of Multimodal Clinical Imaging, and since 2011 a fellow of Jesus College Cambridge. Her current research interests focus on variational methods and partial differential equations for image analysis, image processing and inverse imaging problems. Her research has been acknowledged by scientific prizes, among them the LMS Whitehead Prize 2016, and by invitations to give plenary lectures at several renowned applied mathematics conference, among them the SIAM conference on Imaging Science in 2014, the SIAM conference on Partial Differential Equations in 2015, the IMA Conference on Challenges of Big Data in 2016 and the SIAM annual meeting in 2017.
Carola graduated from the Institute for Mathematics, University of Salzburg (Austria) in 2004. From 2004 to 2005 she held a teaching position in Salzburg. She received her PhD degree from the University of Cambridge in 2009. After one year of postdoctoral activity at the University of Göttingen (Germany), she became a Lecturer in at DAMTP in 2010, promoted to Reader in 2015.

Research

She is interested in the interaction of mathematical sciences and imaging. She studies non-smooth and possibly non-convex variational methods and nonlinear partial differential equations for image analysis and inverse imaging problems, among them image reconstruction and restoration, object segmentation, and dynamic image reconstruction and analysis such as fast flow imaging, object tracking and motion analysis in videos. Moreover, she works on computational methods for large-scale and high-dimensional problems appearing in, e.g. image classification and 3D and 4D imaging.
Within this context she is interested in both the rigorous theoretical and computational analysis of the problems considered as well as their practical implementation and their use for real-world applications.
Currently, her research focuses on customising variational image analysis and image reconstruction models to applications by learning their setup from real-world data training sets. To this end she investigates so-called bilevel optimisation techniques in which the solution is typically constrained to a non-smooth variational problem or a nonlinear PDE.
She has active interdisciplinary collaborations with clinicians, biologists and physicists on biomedical imaging topics, chemical engineers and plant scientists on image sensing, as well as collaborations with artists and art conservators on digital art restoration.

Edinburgh

Dr Kenneth Heafield

Edinburgh

Bio

Kenneth Heafield is a Lecturer in the School of Informatics at the University of Edinburgh where he leads a machine translation group.  He wrote the KenLM library for efficient n-gram language modeling and now works to make neural machine translation faster and higher-quality.  He holds a PhD from Carnegie Mellon’s School of Computer Science and did a postdoc at Stanford.

Research

Kenneth combines natural language processing and systems to make and scale models for machine translation.  Neural networks have improved translation quality and their ability to learn, or not learn, natural language phenomena is an open research area.  However, the high computational cost of training neural network models has slowed experimentation and has forced researchers to decrease training data sizes, sometimes by three orders of magnitude.  Through the Alan Turing Institute, I am collaborating with Intel to accelerate neural networks and challenge the HPC community with real-world natural language processing tasks.

Professor Chris Williams (ULD, Edinburgh)

Edinburgh

Bio

Chris Williams is Professor of Machine Learning in the School of Informatics, University of Edinburgh, and is The Alan Turing Institute’s University Liaison Director for Edinburgh. He obtained his MSc (1990) and PhD (1994) at the University of Toronto, under the supervision of Geoff Hinton. He was a member of the Neural Computing Research Group at Aston University from 1994 to 1998, and has been at the University of Edinburgh since 1998.

Research

Chris is interested in a wide range of theoretical and practical issues in machine learning, statistical pattern recognition, probabilistic graphical models and computer vision. This includes theoretical foundations, the development of new models and algorithms, and applications. His main areas of research are in models for understanding time-series, visual object recognition and image understanding, unsupervised learning, and Gaussian processes. At the Turing he also has interests in improving the data analytics process, looking to address the issues of data understanding and preparation that are widely quoted as taking around 80% of the time in a typical data mining project.

Dr Aretha Teckentrup

Edinburgh

Bio

Dr Teckentrup is a lecturer in data science in the School of Mathematics. Before coming to Edinburgh, she held postdoctoral research positions at Florida State University and the University of Warwick. Dr Teckentrup received her PhD in applied mathematics from the University of Bath in 2013.

Research

Aretha’s research interests are at the interface of numerical analysis, statistics and data science. She is particularly interested in uncertainty quantification in simulation with complex computer models, with recent research focussing on multilevel sampling methods, Bayesian inverse problems, Gaussian process regression, approximation theory and deep Gaussian processes.

Dr. Kostas Zygalakis

Edinburgh

Bio

He received his PhD in computational stochastic differential equations from University of Warwick at 2009 and held postdoctoral positions at the Universities of Cambridge, Oxford and the Swiss Federal Institute of Technology, Lausanne. In 2011 he was awarded a Leslie Fox Prize (IMA UK) in numerical analysis. Before joining Edinburgh as a lecturer in the mathematics of Data Science in 2016, he held a lectureship in Applied Mathematics at the University of Southampton.

 

Research

His research is on the interplay between numerical analysis and computational statistics. In particular, he is interested in the properties of long time approximation to stochastic differential equations and their connection to sampling algorithms. Furthermore, he has recently started working on problems related to Data Assimilation and Bayesian inverse problems, as well as applications of ideas of numerical analysis to machine learning algorithms.

Professor Peter Boyle

Edinburgh

Professor Ruth King

Edinburgh

Bio

Ruth King is the Thomas Bayesí Chair of Statistics at the University of Edinburgh. She was awarded her PhD in 2001 from the University of Bristol. She then held positions at the Universities of Cambridge (PDRA; 2001-3) and St Andrews (lecturer 2003-10; reader 2010-15) before taking up her appointment at the University of Edinburgh in 2015.

Research

My main research interests involve the development of new statistical models and associated model-fitting techniques and their application to real data problems. Particular areas of interest include hidden Markov models/state-space models; missing data; Bayesian inference; incorporating different forms of heterogeneity; capture-recapture; and applications particularly within ecology and epidemiology.

Dr Nick Polydorides

Edinburgh

Bio

Nick Polydorides is a Senior Lecturer in the Institute for Digital Communications, at the School of Engineering at Edinburgh University. He was joined Edinburgh in fall 2013 from the MIT-established Cyprus Institute where he was an Assistant Professor. Prior to that he held postdoctoral positions at the University of Manchester and the MIT. His research interests are in applied inverse problems and in particular algorithms for large-scale electromagnetic imaging and tomography in geophysics, biomedicine, and material non-destructive characterisation.

Research

At the Alan Turing Institute I am keen to explore new algorithms for ëfitting data into modelsí in the context of Bayesian inference. Some key questions underpinning this quest are: how can we predict efficiently and accurately model observations conditioned to our prior knowledge on the models? Reversely, how can model parameters be estimated with robustness from incomplete and inconsistent data sets? Relevant to this context is also the problem of optimal design, which aims to yield an optimal set of data whose information n content complements that of the prior models, and yield estimators with suppressed uncertainty levels. I hope to address some of these questions, and implement some new ideas in code and analysis.

Dr Beatrice Alex

Edinburgh

Bio

Dr. Beatrice Alex is a Senior Research Fellow at the School of Informatics at the University of Edinburgh.  Her research interests center around information extraction and text mining.  She has worked on a many projects involving natural language processing applied to different domains, including biomedicine, astronomy, news, recruitment, history and most recently social media and literature.

Research

Project Description:
Title: Data Analytics for Large Speech Archives: Mining transcripts of audio and video collections
Accessibility, interpretability and linking of data have become the focus of many data providers and users. Most research in this area has been directed towards making textual resources more accessible. Audio and video archives need to be regarded as equally important. Cataloguing and transcription of speech archives is extremely labour-intensive and time-consuming. The goal of my proposal is to mine transcripts of large speech archives automatically to generate metadata for such collections. I will be working closely with the British Library and apply my work to TV and radio recordings held at their archive.

Professor David Aspinall

Edinburgh

BIO

David Aspinall holds a personal chair in Software Safety and Security at University of Edinburgh. He leads the Security and Privacy research group in the School of Informatics and a cross-discipline Cyber Security & Privacy Research Network in the University.  His research interests cover foundational and applied areas of computer security, as well as interactive theorem proving and proof engineering.

 

RESEARCH

Professor Aspinall wants to pursue several ideas in the Alan Turing Institute.  First, he plans to work on applying interactive theorem proving to cryptography, by formalising game and simulation based cryptographic soundness proofs on machine.  Second, he wants to build on earlier work that combines machine learning and formal methods, which he used to construct robust malware classifiers for Android applications: he wants to extend this work to other sources of data, including data collected by the security industry, such as network traffic and log data, to develop better methods of anomaly detection and inference from event streams.  Finally, he is interested in privacy and security issues behind personal data, including data that is collected in eHealth and mHealth solutions, or IoT devices.

 

Dr Natalia Bochkina

Edinburgh

BIO

Natalia Bochkina joined the University of Edinburgh as a Lecturer in Statistics in 2007. In 2003-2007 she was a postdoc at the Biostatistics group at the Imperial College London working on the collaborative project building a biological atlas of insulin resistance. In 2002-2003 she was a statistician in biopharmaceutical company Oxford GlycoSciences (Ltd), completing her PhD in 2002 at the University of Bristol.

 

RESEARCH

I am interested in understanding theoretical properties of decision-making under uncertainty which arises in statistics and in machine learning motivated by practical applications. Currently my main areas of research are study of the rate of convergence and local approximation of posterior distribution (Bernstein-von Mises theorem) in Bayesian nonparametric, semi-parametric and high dimensional models which can be misspecified, non-regular or ill-posed, including inverse problems.

In additional to theory, I am interested in statistical modelling of high throughput genomic data, including the current project on estimating gene interaction networks using graphical models.

An exciting emerging area is taking into the account computational complexity when studying efficiency of the decision-making. One of the tools is calibration of faster approximate Bayesian models using techniques for misspecified models to achieve best possible efficiency.

Professor Anthony Kennedy

Edinburgh

BIO

Tony Kennedy has been Professor of Computational Science in the School of Physics and Astronomy at the University of Edinburgh since 1998, before which he spent 19 years in the United States working at the University of Maryland, the Institute for Theoretical Physics (ITP) at the University of California at Santa Barbara (UCSB), and the Supercomputer Computations Research Institute (SCRI) at Florida State University (FSU).

 

RESEARCH

His research interests span several disciplines within Theoretical Physics, Mathematics, and Computer Science.

  • His principal research area has been developing algorithms for non-perturbative computations in relativistic quantum field theory, particularly in Quantum Chromodynamics (the theory of the strong nuclear force).  This involves evaluating infinite-dimensional (functional) integrals numerically using Markov Chain Monte Carlo (MCMC) methods.  He introduced the Hybrid (or Hamiltonian) Monte Carlo algorithm in 1987, and several subsequent developments based on this such as the Rational HMC algorithm (which uses optimal rational approximations to improve stochastic estimates for functional determinants), and methods for using Shadow Hamiltonians to automate the determination of optimal parameters for symplectic integrators for use in HMC computations.

 

He is especially interested in applying MCMC and other statistical methods to other areas, such as machine learning (ML), as well as developing a better understanding the theoretical basis of ML in general.

  • He has also interested in high-performance computer architecture and in ways of developing efficient, modular, reusable software for such systems. He was part of the collaboration that built the QCDSP computer for lattice QCD computations (which won a Gordon Bell prize in 1998), and in the subsequent QCDOC computer which directly led to IBM’s Blue Gene supercomputers.
  • Other interests include various areas of theoretical and mathematical physics (such as renormalization theory), algorithms for computations using representations of groups and algebras, algorithms for computer algebra and related category theoretic ideas for computer languages.

 

Professor Ben Leimkuhler

Edinburgh

Bio

I study the design, analysis and implementation of algorithms for time-dependent phenomena and modelling for problems in engineering and the sciences. My previous works have helped to establish the foundations of molecular simulation, providing efficient deterministic and stochastic numerical methods for an exploding field of application. A recent line of research as focussed on stochastic algorithms for Bayesian inference from data and has demonstrated the potential for crossover of work from molecular science to data analytics (in this case molecular dynamics temperature controls were used to stabilise sampling-based parameterisation schemes).

Research

Data science disrupts the traditional mathematical model by replacing physical law with empirical law, through the need to incorporate inference based on massive data sets or streams, and by destroying smooth structures (ODE/PDE solutions) that underpin numerical analysis.   Data science is grounded in statistics and optimization, but to be effective in the engineering setting, data analysis methods such as Bayesian inference must be merged with models built up over many centuries on a foundation of physical law (e.g. quantum mechanics and thermodynamics) and be compatible with dynamical principles and geometric (or topological) constraints. During my fellowship I plan to explore the interplay between “naive” data science approaches and models informed by physical law and mathematical structure. Access to the Alan Turing Institute is already providing me with connections to the statistics, operational research and machine learning communities. I look forward to expanding these connections in the coming years.

Dr Ioannis Papastathopoulos

Edinburgh

BIO

I was born in Athens, Greece, and received my undergraduate training in Statistics and Insurance Science at the University of Piraeus. Then I received my MSc in Statistics and PhD in Statistics from Lancaster University and held a Brunel Research Fellowship in Statistics at the School of Mathematics, University of Bristol.

 

RESEARCH

My ATI research related interests and activities focus on the broad area of extreme value modelling and inference for rare and catastrophic events.Extreme events are multi-dimensional in nature, can cause havoc for the people affected and typically result in large financial losses. A wide variety of geophysical data is now acquired with very broad wavelength ranges from satellites and many other sources.This has led to increasingly complex and large datasets which require new methodological tools to analyse.I am particularly
interested in exploring novel approaches for developing such tools and in identifying and communicating risks imposed by the society, economy and nature.My research interests include:

– Conditional dependence structures and Graphical models;
– item High-dimensional inference;
– item Scalable data fitting algorithms;
– item Spatio-temporal extreme value modelling.

Dr Peter Richtarik

Edinburgh

BIO

Peter Richtarik is a Reader in the School of Mathematics at the University of Edinburgh, and is the Head of a Big Data Optimization Lab. He received his PhD from Cornell University in 2007, and currently holds an EPSRC Early Career Fellowship in Mathematical Sciences.

 

RESEARCH

My main research focus is the development of new optimization algorithms and theory. In particular, much of my recent work is in the emerging field of big data optimization, with applications in machine learning in general and empirical risk minimization in particular. For big data optimization problems, traditional methods are no longer suitable, and hence there is need to develop new algorithmic paradigms. An important role in this respect is played by randomized algorithms of various flavors, including randomized coordinate descent, stochastic gradient descent, randomized subspace descent and randomized quasi-Newton methods. Parallel and distributed variants are of particular importance.

Dr Lukasz Szpruch

Edinburgh

BIO

I am a Lecturer at the School of Mathematics, University of Edinburgh. Before moving to Scotland I was a Nomura Junior Research Fellow at the Institute of Mathematics, University of Oxford, and a member of Oxford-Man Institute for Quantitative Finance. I hold a Ph.D. in mathematics from University of Strathclyde in Glasgow.

 

RESEARCH

My research interests are in theoretical and applied probability theory. I have a particular interest in:

– (Multilevel) Monte Carlo Methods
– Stochastic Interacting Particle Systems and Mean Filed Limits; Stochastic McKean-Vlasov Equations
– Sequential Monte Carlo (Particle Filters)
– (Stochastic) Gradient type algorithms and their applications in statistics and optimization

– Highly dimensional stochastic systems such as SDEs, BSDEs and SPDEs

Dr Charles Sutton

Edinburgh

Bio

Charles Sutton is a Reader (equivalent to Associate Professor: http://bit.ly/1W9UhqT) in Machine Learning at the University of Edinburgh. He has over 50 publications in a broad range of applications of probabilistic machine learning. His work in machine learning for software engineering has won an ACM Distinguished Paper Award. His PhD is from the University of Massachusetts Amherst, and he has done postdoctoral work at the University of California Berkeley. He is currently Director of the EPSRC Centre for Doctoral Training in Data Science at the University of Edinburgh.

Research

My research focuses on developing new machine learning methods that are motivated by the demands of new, cutting-edge practical problems. I develop new techniques in probabilistic machine learning, approximate inference in graphical models, and more recently deep learning.
In general, these motivating applications come from a wide range of areas, including natural language processing, analysis of computer systems, software engineering, sustainable energy, and exploratory data analysis. Most recently, I am particularly fascinated by two important application areas: machine learning and NLP methods to improve software development, and machine learning and artificial intelligence methods to support the full practical workflow of data science.

Oxford

Professor Jared Tanner (ULD, Oxford)

Oxford

Bio

Jared Tanner is Professor of the Mathematics of Information at the University of Oxford and The Alan Turing Institute’s University Liaison Director for Oxford. He obtained his PhD (2002) in applied mathematics at the University of California at Los Angeles, and was a postdoctoral fellow at the University of California at Davis (Maths) and Stanford University (Stats.) where he worked with David L. Donoho.  Prior to joining the University of Oxford in 2012 he was Professor of the Mathematics of Information at the University of Edinburgh (2007-2012).  He is founding editor-in-chief of Information and Inference: A Journal of the IMA, whose mission is to publish high quality mathematically oriented articles furthering the understanding of the theory, methods of analysis, and algorithms for information and data.  He is also on the editorial board for Applied and Computational Harmonic Analysis, Multiscale modelling and simulation A SIAM Interdisciplinary Journal, and was an associate editor for the Princeton Companion to Applied Mathematics.  His research has appeared in the Proc Natl Acad Sci USA, Phil Trans Royal Soc A, and other leading journals.

Research

Jared Tanner’s research concerns extracting models of high dimensional date which reveal of the essential information in the data.  Specific contributions include the derivation of sampling theorems in compressed sensing using techniques from stochastic geometry and the design and analysis of efficient algorithms for matrix completion which minimise over higher dimensional subspaces as the reliability of the data warrants.  These techniques allow more efficient information acquisition as well as the ability to cope with missing data.

Recent interests include new models for low dimensional structure in heterogeneous data and topological data analysis.

Professor Stephen Roberts

Oxford

BIO

Stephen Roberts is the RAEng/Man Professor of Machine Learning at the University of Oxford. Stephen is a Fellow of the Royal Academy of Engineering, the Royal Statistical Society, the IET and the Institute of Physics. Stephen is Director of the Oxford-Man Institute of Quantitative Finance and Director of the Oxford Centre for Doctoral Training in Autonomous Intelligent Machines and Systems (AIMS).

 

RESEARCH

Stephen’s interests lie in methods for machine learning & data analysis in complex problems, especially those in which noise and uncertainty abound. His current major interests include the application of machine learning to huge astrophysical data sets (for discovering exo-planets, pulsars and cosmological models), biodiversity monitoring (for detecting changes in ecology and spread of disease), smart networks (for reducing energy consumption and impact), sensor networks (to better acquire and model complex events) and finance (to provide timeseries and point process models and aggregate large numbers of information streams).

Professor Alessandro Abate

Oxford

Bio

Alessandro Abate is an Associate Professor in the Department of Computer Science at the University of Oxford. He received a Laurea in 2002 from the University of Padova, an MS in 2004 and a PhD in 2007 at UC Berkeley (USA). He has been an International Fellow at SRI International in Menlo Park (USA), and a PostDoctoral Researcher at Stanford University (USA). From 2009 to 2013 he has been an Assistant Professor at TU Delft (NL).

Research

His research interests lie on the analysis, formal verification, and optimal control of stochastic hybrid systems, which are complex and heterogeneous dynamical models encompassing discrete and continuous dynamics, and with dynamics enriched by randomness. Beyond research on deductive, model-based techniques, he works on data-driven (inductive) approaches for model construction and estimation, with techniques from computational statistics (machine learning) and systems identification (an area within systems and control theory). The overall plan is to fully and precisely integrate data-driven procedures within model-based verification architectures.

Professor Pawan Kumar

Oxford

Bio

M. Pawan Kumar is an Associate Professor of Information Engineering at the University of Oxford and a fellow of Lady Margaret Hall. He is a principal researcher in the OVAL group, which focuses on the design and analysis of discrete and continuous optimization algorithms for the problems encountered in computer vision and machine learning.

Research

At the Turing, he plans to explore various problems that are central to scaling up machine learning for visual data. Examples include efficient and principled optimization algorithms for deep neural networks, learning from large-scale weakly supervised data such as freely downloadable images or videos on the Internet that have sparse and often noisy tags, compressing neural networks for memory and speed efficiency, and fast combinatorial algorithms and relaxation-based approaches for structured prediction.

Professor Varun Kanade

Oxford

Bio

Varun Kanade is a Research Lecturer in the Department of Computer Science, Oxford University. He has been a Simons Postdoctoral Fellow at the University of California, Berkeley and a FSMP postdoctoral fellow at ENS, Paris. He obtained his Ph.D. from Harvard University in 2012. His research interests are in machine learning and theoretical computer science.

Research

Varun Kanade’s research interests lie at the intersection of machine learning and theoretical computer science. At the ATI, his research will focus on mathematical foundations of machine learning, in particular statistical and computational tradeoffs, as well as improving our theoretical understanding of methods successful in practice. His work will also focus on randomized algorithms particularly for analysing large networks.

Professor Yee Whye Teh

Oxford

BIO

Yee Whye Teh is a Professor of Statistical Machine Learning at the University of Oxford, an Alan Turing Institute Faculty Fellow and a Research Scientist at Google DeepMind.  He obtained his Ph.D. at the University of Toronto, and did postdoctoral work at the University of California at Berkeley and National University of Singapore (as Lee Kuan Yew Postdoctoral Fellow).  He was a Lecturer then a Reader at the Gatsby Computational Neuroscience Unit, UCL prior to his current appointment.

 

RESEARCH

At the Alan Turing Institute, he is interested in investigating novel scalable machine learning and computational statistics methods, in particular in the frameworks of probabilistic models, Bayesian nonparametrics and deep learning. He is interested in applications to machine perception, recommender systems, genetics and genomics.

Professor Coralia Cartis

Oxford

BIO

Coralia Cartis (BSc Mathematics, Babesh-Bolyai University, Romania; PhD Mathematics, University of Cambridge (2005)) has joined the Mathematical Institute at Oxford and Balliol College in 2013 as Associate Professor in Numerical Optimization. Previously, she worked as a research scientist at Rutherford Appleton Laboratory and as a postdoctoral researcher at Oxford University. Between 2007-2013, she was a (permanent) lecturer in the School of Mathematics, University of Edinburgh.

 

RESEARCH

Assoc. Prof. Cartis’ research is on optimization, algorithm development, analysis and implementation for a variety of problem classes (linear, convex, nonconvex, smooth/nonsmooth, stochastic), suitable for large scale. In the past few years, her research focused on complexity of nonconvex optimization problems; compressed sensing; parameter estimation for climate modelling. Currently, she is investigating optimization methods for problems with imperfect or corrupted information, such as in the case of stochastic optimisation.

Professor Raphael Hauser

Oxford

Bio

Raphael Hauser studied Mathematics and Theoretical Physics at the EPFL and ETH in Lausanne and Zurich, Switzerland, followed by a PhD in Operations Research at Cornell University in Ithaca, USA. After a postdoc at Cambridge, Raphael joined the faculty at the University of Oxford, where he is currently an Associate Professor in Numerical Mathematics at the Oxford Mathematical Institute and the Tanaka Fellow in Applied Mathematics at Pembroke College.
He is a member of the Numerical Analysis Group, the Mathematical and Computational Finance Group, and of the Oxford Emirates Data Lab.

Research

Raphaelís research interests lie in optimisation algorithms, applied probability and statistics, distributed computing, and medical imaging. He is experienced in working on industrial collaborations involving real-world data and an inventor of several patents. At the Alan Turing Institute he will work on distributed algorithms for convex optimisation, extending an approach he developed for large scale principal component analysis.

Professor Francois Caron

Oxford

BIO

Since September 2015, Professor Francois Caron has been an Associate Professor in the department of Statistics and a tutorial Fellow at Keble College, University of Oxford. From September 2013 to August 2015, he was a Marie Curie Research Fellow in the department of Statistics at the University of Oxford. From 2008 to 2013, he was a Research Scientist (Chargé de Recherche) at INRIA Bordeaux. He completed his Ph.D. in Information Engineering at the University of Lille I in 2006, and spent two years as a post-doctoral researcher in the departments of Computer Science and Statistics at the University of British Columbia (2006-2008).

 

RESEARCH

Prof. François Caron’s research interests lie in the development of statistical models and computational procedures for the analysis of structured data, with particular interest in Bayesian nonparametrics and Monte Carlo methods. His recent research focuses on building probabilistic models of network data which can capture the salient properties of real-world networks (sparsity, modularity), as well as efficient algorithms for learning the hidden structure of those networks.

Professor Dino Sejdinovic

Oxford

BIO

Dino Sejdinovic is an Associate Professor at the Department of Statistics, University of Oxford and a Fellow of Mansfield College. He previously held postdoctoral positions at the Gatsby Computational Neuroscience Unit, University College London (2011-2014) and at the Statistics Group, University of Bristol (2009-2011). He received a PhD in Electrical and Electronic Engineering (Bristol, 2009).

 

RESEARCH

Dino is broadly interested in statistical foundations underpinning large-scale machine learning algorithms, with a particular emphasis on nonparametric and kernel methods. Recent research focused on discovery of higher-order interactions in datasets (when weak individual causes combine in a nonlinear way to form a strong effect), as well as on adaptive inference suited for complex models with nonlinear dependencies and intractable likelihoods. Further interests include tradeoffs between statistical, computational and communication properties of learning algorithms.

Professor Thomas Lukasiewicz

Oxford

Professor Arnaud Doucet

Oxford

BIO

He obtained his PhD from the University of Paris XI (Orsay) in 1997. He previously held academic positions at Cambridge University, Melbourne University, The Institute of Statistical Mathematics in Tokyo and the University of British Columbia where he was a Canada Research Chair in Stochastic Computation.

RESEARCH

Prof. Arnaud Doucet’s research concerns numerical methods for the analysis of complex data sets. In particular, he has contributed to the development and study of sequential Monte Carlo and Markov chain Monte Carlo methods.

Professor Chris Holmes

Oxford

BIO

He is a Professor of Biostatistics at the University of Oxford with a joint appointment between the Department of Statistics and the Nuffield Department of Clinical Medicine through the Wellcome Trust Centre for Human Genetics. He holds a Programme Leader’s award in Statistical Genomics from the Medical Research Council UK.

 

RESEARCH

My research interests are in statistical methods and statistical machine learning approaches for problems arising in modern biomedical science and population health. In particular in the theory, methods and applications of probabilistic data modelling and the use of subjective probability theory as a unified framework for coherent inference to tackle highly structured yet heterogeneous data sets that are now being routinely collected in these application domains. This has lead him to investigate Bayesian methods and computation, particularly for nonlinear systems in the presence of model misspecification.

Professor Terry Lyons

Oxford

BIO

Professor Terry Lyons is the Wallis Professor of Mathematics; he was a founding member (2007) of, and then Director (2011-2015) of, the Oxford Man Institute of Quantitative Finance; he was the Director of the Wales Institute of Mathematical and Computational Sciences (WIMCS; 2008-2011). He came to Oxford in 2000 having previously been Professor of Mathematics at Imperial College London (1993-2000), and before that he held the Colin Maclaurin Chair at Edinburgh (1985-93).

 

RESEARCH

Prof Lyons’ long-term research interests are all focused on Rough Paths, Stochastic Analysis, and applications – particularly to Finance and more generally to the summarsing of large complex data. More specifically, his interests are in developing mathematical tools that can be used to effectively model and describe high dimensional systems that exhibit randomness. Prof Lyons is involved in a wide range of problems from pure mathematical ones to questions of efficient numerical calculation.

Professor Ulrike Tillmann

Oxford

BIO

Prof. Ulrike Tillmann FRS has been at the University of Oxford since 1992. She is an algebraic topologist, known in particular for her work on Riemann surfaces and the homology of their moduli spaces. She has long standing research interest in homology stability questions. In 2011 she introduced an annual course (with Abramsky) in Computational Algebraic Topology at masters level. In the last year, Tillmann has co-organized four workshops on topological data analysis, as well as an CMI-LMS research school.

She held an EPSRC Advanced Fellowship 1997-2003. She was invited to present at the ICM in 2002 and was a member of the topology subject panel for both the 2010 and 2014 ICMs. In 2008 she was made a Fellow of the Royal Society and received the Bessel Forschungspreis from the Humboldt Gesellschaft. She is an inaugural Fellow of the AMS.

 

RESEARCH

Algebraic topology and its applications. Algebraic topology is a very effective tool to study the global properties of geometric objects. For example, take the surface of a ball and divide it into triangles; now count the number of faces, add the number of vertices and subtract the number of edges; no matter how you choose your triangles, the result will always be 2. Do the same with the surface of a donut and the result is always 0. These numbers were already known by Euler and are foreshadows of homology developed in the 20th century. By now the basic ideas of algebraic topology have permeated nearly every branch of research in mathematics.

Her own research has been motivated by questions in quantum physics and string theory. In particular, she has contributed to our understanding of the ‘space of surfaces’.

Professor Frank Wood

Oxford

BIO

Prof. Wood is an associate professor at the Department of Engineering Science, University of Oxford. Before that Dr. Wood was an assistant professor of Statistics at Columbia University and a research scientist at the Columbia Center for Computational Learning Systems. He formerly was a postdoctoral fellow of the Gatsby Computational Neuroscience Unit of the University College London under Dr. Yee Whye Teh. He received his PhD from Brown University in computer science under the supervision of Dr. Michael Black and Dr. Tom Griffiths.

Prof. Wood is a product of the Illinois Mathematics and Science Academy (1992). He began college at the University of Illinois at Chicago (UIC) but transferred and received a B.S. in Computer Science from Cornell University (1996). Prior to his academic career he was a successful entrepreneur having run and sold the content-based image retrieval company ToFish! to Time Warner and serving as CEO of Interfolio. He started his career working at both the Cornell Theory Center and subsequently the Lawrence Berkeley National Laboratory.

Professor Ian Horrocks

Oxford

BIO

Ian Horrocks is a Professor of Computer Science at the University of Oxford and a Visiting Professor in the Department of Informatics at the University of Oslo. He is a Fellow of the Royal Society, a member of Academia Europaea, an ECCAI Fellow and a Fellow of the British Computer Society.

 

RESEARCH

His research interests include logic-based knowledge representation and reasoning and semantic technologies, with a particular focus on ontology languages and applications. He was an author of the OIL, DAML+OIL, and OWL ontology language standards, chaired the W3C working group that standardised OWL 2, and developed many of the algorithms, optimisation techniques and reasoning systems that underpin OWL applications. His recent work includes query answering over ontologies and very large data sets, and applications in domains such as engineering, oil and gas, finance and medicine.

Professor Dan Olteanu

Oxford

BIO

Dan Olteanu is an associate professor in the Department of Computer Science and Fellow of St Cross College at the University of Oxford. He has also taught at University of California, Berkeley, University of Munich, Saarland University, and Heidelberg University. He received his PhD in Computer Science from University in Munich in 2005.

 

RESEARCH

His research interests are in database systems and theory. Dan contributed to XML query processing, incomplete information and probabilistic databases, and more recently to factorized databases and the industrial-strength LogicBlox database and analytics system. He is a co-author of the book “Probabilistic Databases” (2011). Olteanu has served as associate editor for PVLDB’13 and IEEE TKDE, as track chair for ICDE’15, group leader for SIGMOD’15, and will serve as vice chair for SIGMOD’17. His current research is supported by awards from Amazon, Google, and LogicBlox, and an ERC consolidator grant.

 

 

Professor Shimon Whiteson

Oxford

BIO

Professor Shimon Whiteson studied English and Computer Science at Rice University before completing my doctorate in Computer Science at the University of Texas at Austin in 2007.  He then spent eight years as an Assistant and then an Associate Professor at the University of Amsterdam before joining Oxford as an Associate Professor in 2015.  He was awarded an ERC Starting Grant from the European Research Council in 2014.

 

RESEARCH

Assoc. Prof. Shimon Whiteson’s research focuses on artificial intelligence. His goal is to design, analyse, and evaluate the algorithms that enable computational systems to acquire and execute intelligent behaviour. He’s particularly interested in machine learning, with which computers can learn from experience, and decision-theoretic planning, with which they can reason about their goals and deduce behavioural strategies that maximise their utility.

In addition to theoretical work on these topics, he has in recent years also focused on applying them to practical problems in robotics and search engine optimisation.

Dr Sarah Filippi

Oxford

BIO

Dr. Sarah Filippi joined the Department of Statistics of Oxford University in June 2014. She previously held a Medical Research Council Fellowship (2011-2104) in the Theoretical Systems Biology group at Imperial College London where she worked on a range of topics in computational statistics focused on understanding biological processes and their relation to disease. Prior to this she studied mathematics and stochastic processes at the University Denis Diderot in Paris (France) and completed her PhD in 2010 in reinforcement learning and parametric bandit models at LTCI, a joint lab of TELECOM ParisTech and CNRS.

 

RESEARCH

Dr. Filippi’s main research interests are related to the use of mathematical modelling, statistical machine learning and computational statistics to gain insight into biological processes and their role in diseases. She has a particular interest in addressing computational and statistical challenges in the analysis of genomic, transcriptomic and proteomic data at a single-cell level.

Professor Luciano Floridi

Oxford

BIO

Luciano Floridi is Professor of Philosophy and Ethics of Information at the University of Oxford, where he is the Director of Research of the Oxford Internet Institute, Governing Body Fellow of St Cross College, Distinguished Research Fellow of the Uehiro Centre for Practical Ethics in the Faculty of Philosophy, and Research Associate and Fellow in Information Policy of the Department of Computer Science. He is also Adjunct Professor of the Department of Economics, American University, Washington D.C. His most recent book is The Fourth Revolution – How the infosphere is reshaping human reality (Oxford University Press, 2014 and 2016). He is Member of the EU Ethics Advisory Group, Chairman of Ethics Advisory Board of the European Medical Information Framework, Member of Google’s Advisory Board on “the right to be forgotten” and Member of the Advisory Board of Tencent’s Internet and Society Institute.

 

RESEARCH

His areas of research are: philosophy of information, information ethics (including data ethics and computer ethics), philosophy of AI, philosophy of technology, epistemology, and logic.
http://www.oii.ox.ac.uk/people/floridi/

Dr Mariarosaria Taddeo

Oxford

BIO

Dr. Mariarosaria Taddeo Researcher – Oxford Internet Institute, University of Oxford and Faculty Fellow at the Alan Turing Institute. Her recent work focuses mainly on the ethical analysis of cyber security practices and information conflicts.

Dr. Taddeo is the 2010 recipient of the Simon Award for Outstanding Research in Computing and Philosophy and of the 2013 World Technology Award for Ethics. She is Co-Investigator, PETRAS Hub for Internet of Things, a EPSRC project and serves editor-in-chief of Minds & Machines, in the executive editorial board of Philosophy & Technology, and is the President of the International Association of Computing and Philosophy.

 

RESEARCH

Her area of expertise is Information and Computer Ethics, although she has worked on issues concerning Philosophy of Information, Epistemology, and Philosophy of AI. She published several papers focusing on online trust, cyber security and cyber warfare and guest-edited a number of special issues of peer-reviewed international journals: Ethics and Information Technology, Knowledge, Technology and Policy, Philosophy & Technology. She also edited (with L. Floridi) a volume on ‘The Ethics of Information Warfare’ (Springer, 2014) and is currently writing a book on ‘The Ethics of Cyber Conflicts’ under contract for Routledge.

Professor Charlotte Deane

Oxford

BIO

Charlotte has a wide range of research interests concerning protein structure prediction and protein interaction networks, combining both theoretical work and empirical analyses. Originally a Chemist she has moved across many traditional research disciplines and her research group is now based in Statistics. Charlotte completed her PhD in Cambridge and after a Wellcome trust fellowship in California she returned to Oxford as a University Lecturer in 2002.

 

RESEARCH

Prof. Charlotte Deane’s group works in the protein bioinformatics area. Currently the research focuses on understanding protein structure and improving our ability to model and design proteins.

To this end, the group is developing a rigorous definition of evolutionary relationships between proteins of differing structures using, in the initial stages, genomic data and building novel protein structure prediction software.

In a separate thread the group has been examining protein-protein interaction networks in terms of both their quality and use as a predictive tool.

UCL

Professor David Pym (ULD, UCL)

UCL

Bio

David Pym is Professor of Information, Logic, and Security at UCL and is The Alan Turing Institute’s University Liaison
Director for UCL. Heholds a PhD in logic and theoretical computer science from Edinburgh,and an MA and an ScD in mathematics from Cambridge. He is a Fellow of the IMA and the BCS. David spent many years with Hewlett-Packard’s Research Laboratories, where he developed interests in systems, security, and economics.

Research

David will work on a range of topics in security and privacy in distributed information-processing systems. He is interested in questions about access control in distributed systens, security and privacy policy, and the economics of security management. He is also interested in understanding basic questions in distributed systems architecture and behaviour, such as consistency and the relationship between systems management policies and systems architecture. David addresses these issues using ideas and techniques from logic, theoretical computer science, probability theory, and economics. He aims to build both conceptual and implemented tools to support decision-making in systems and policy design.

Dr Emine Yilmaz

UCL

BIO

She is a senior lecturer (associate professor) at University College London, Department of Computer Science and a faculty fellow of the Alan Turing Institute. She also works as a research consultant for Microsoft Research, Cambridge and serve as one of the organizers of the Centre for Computational Statistics and Machine Learning (CSML) at UCL. She is the recipient of the Karen Sparck Jones Award in 2015 and is one of the recipients of the Google Faculty Research Award in 2014/2015.

 

RESEARCH

Her research interests lie in the areas of information retrieval, web science, and applications of machine learning, probability and statistics. For more information about recent publications, please visit my publications page.

Dr Ricardo Silva

UCL

BIO

Ricardo Silva got his PhD at Carnegie Mellon in 2005, in the newly formed Machine Learning Department. He moved to UCL as a Senior Research Fellow in the Gatsby Computational Neuroscience Unit. After a year at the Statistical Laboratory at Cambridge as a postdoc, Ricardo returned to UCL to join the Department of Statistical Science as a Lecturer in 2008.

 

RESEARCH

His research focuses on:

1. Algorithms for probabilistic inference: approximations for likelihoods and posterior distributions based mostly of variational approximations and Markov chain Monte Carlo.

2. Latent variable models:  measurement error models and generalisations of probabilistic principal component analysis, as well as the modelling of network data.

3. Machine learning for causal inference: identification and discovery of models with unmeasured confounding and measurement error.

He has also recently focused on applied work on human movement modelling, including partners such as TfL (for traffic data), UCL Institute of Behavioural Neuroscience (human navigation strategies) and Stratagem Ltd (sports modelling).

Dr David Barber

UCL

BIO

He studied mathematics as an undergraduate (Cambridge) and the Statistical Mechanics of Neural Nets for my PhD (Edinburgh). He has worked in various institutions since then, joining the Department of Computer Science at UCL around 2006.

 

RESEARCH

He is interested in developing methods to make progress in tackling fundamental challenges in Artificial Intelligence.  Particularly interesting areas for me are reasoning under uncertainty and how to represent knowledge. How can we best use large scale data and large scale computation to advance AI. What kinds of problems can we not currently solve and how could we try to solve them in the future?

Dr Ioanna Manolopoulou

UCL

Bio

Ioanna has been a Lecturer in Statistics at University College London since 2012. She completed her PhD at the University of Cambridge in 2008 under the guidance of Simon TavarÈ and Steve Brooks. She then joined SAMSI (Statistical and Applied Mathematical Sciences Institute) as Postdoctoral Fellow and Duke University as Visiting Assistant Professor, working with Mike West and Sayan Mukherjee.

Research

Ioanna’s work focuses on developing Bayesian modelling and inferential tools for complex data, involving space and time processes with inhomogeneous features. Her work has been applied to various applications including customer science, ecology, health economics and accounting. Methodologically, one of her directions within the ATI will be to develop new methods for dealing with data with complex dependencies or which have not been collected completely randomly. In terms of applications, her work in retail analytics aims to provide novel customer and product segmentations as well as devise more tailored prediction algorithms. She is also interested in health economics, in particular identifying population-wide treatment effects and studying the effect of uncertainty in public health decision-making.

Dr Franz Kiraly

UCL

BIO

As a practical statistician and machine learner, Franz is interested in creating a data analytics workflow which is empirically solid, quantitative, and useful in the real world, with a focus on predictive modelling.

He is working on what he considers to be two of the most pressing challenges in a practical and data-centric context: namely, how to deal with structured data, such as learning with data samples of series, sequences, matrices, or graphs; and how to quantitatively assess and compare methods against each other, for example whether complicated algorithm X is really better than a random guess.

These are especially relevant in applications where usually the data and the associated scientific questions, and not a single method class is in the focus of interest; current project and collaboration domains include the medical sciences, sports and prevention, geoscience, physics and finance.

 

RESEARCH

Recently, Franz has been doing research on these applications:

Prediction and Prevention of Falls in a Neurological In-Patient Population. Falling, and associated injuries such as hip fracture, are a major strain on health and health resources, especially in the elderly or hospitalized. We are able to predict, with high accuracy in a neurological population, whether a patient is likely to fall during their stay, using only a number connecting test (the Trail making test).

(read more)

Quantification and Prediction in Running Sports. Characterizing the training state of running athletes, and making predictions for race planning and training. We can predict Marathon times with an error in the order of a few minutes, and we are able to accurately summarize an athlete by three characteristic numbers.

(read more)

His current work on data analysis methodology includes:

Non-linear prediction and dimension reduction with series-valued samples. We propose a new learning framework for the situation where the data samples are time series or otherwise sequentially ordered, based on kernels whose features are ordered variants of sample moments.
(read more)

Single-Entry Matrix Completion and Local Matrix Completion. Our new methods can (i) reliably impute or predict single missing entries in a numerical data table, with error bars, and (ii) do so without necessarily reading in all entries in a big data table. They are the first of their kind under the common low-rank assumption.

(read more)

Kernel Learning with Invariances. Encoding known invariances of the data, say sign/mirror symmetry, scaling or phase invariance, efficiently with a kernel (work in progress).

(read more)

Dr Ioannis Kosmidis

UCL

Bio

Ioannis Kosmidis is a Senior Lecturer at the Department of Statistical Science in University College London. Having obtained a BSc in Statistics at the Athens University of Economics and Business in 2004, he was then awarded his PhD in Statistics in 2007 at University of Warwick with a thesis titled “Bias reduction in exponential family nonlinear models“. He then held an appointment as a CRiSM Research Fellow at University of Warwick until 2010. In September 2010, Ioannis joined University College London as a Lecturer in the Department of Statistical Science, and got promoted to Senior Lecturer in 2015.

Research

Ioannis’ theoretical and methodological research focuses on optimal estimation and inference from complex statistical models, penalized likelihood methods and clustering. A particular focus of his work is the development of efficient, in terms of computational complexity and implementation, algorithms for applying the methods he develops to prominent data-analytic scenarios. He is doing extensive work in producing corresponding, well-documented, open-source software that delivers the methodological advances to the data science community and beyond (for example, the brglm, profileModel, betareg and trackeR R packages). Ioannis also actively engages in a range of cross-disciplinary applications (e.g. applications in earthquake engineering, finance, and sport and health analytics), particularly in settings where statistical modelling and the associated algorithms can impact policy making. He is the founder and leader of the Statistics in Sports and Health research group at the University College London, and associate editor for Statistics and Computing and the Journal of Statistical Software.

Detailed, up-to-date information on his research, teaching, enabling and engaging activities can be found on his website and his CV.

Professor Petros Dellaportas

UCL

Bio

Petros Dellaportas is a Professor of Statistical Science at the University College London. Previously he held academic positions in Athens University of Economics and Business, Greece.  He has published in Annals of Statistics, Biometrika, JRSSB, Statistical Science, Journal of Business and Economics Statistics, etc.  He is currently associate editor of Biometrika and Electronic journal of Statistics.

Research

His research interests include Bayesian theory and applications, computational statistics, financial modelling. In particular he currently works in problems of multivariate stochastic volatility, Gaussian processes, Bayesian mixture hierarchical models, modelling of high-frequency financial data, modelling of sports data.

Dr Hao Ni

UCL

Bio

Dr Hao Ni is a senior lecturer in financial mathematics at UCL since September 2016. Prior to this she was a visiting postdoctoral researcher at ICERM and Department of Applied Mathematics at Brown University from 2012/09 to 2013/05 and continued his postdoctoral research at the Oxford-Man Institute of Quantitative Finance until 2016. She finished her D.Phil. in mathematics in 2012 under the supervision of Professor Terry Lyons at University of Oxford.

Research

Her research interests include stochastic analysis, financial mathematics and machine learning. More specifically she is interested in non-parametric modelling effects of data streams through rough paths theory and statistical models. Rough paths theory is a non-linear extension of classical theory of control differential equations to model highly oscillatory systems, and the core concept in rough paths theory is the signature of a path, which can be used as useful features for learning to summarize sequential data in terms of its effect. Moreover, she is also interested in its applications, e.g. online Chinese handwritten character and financial data streams.

Dr George Danezis

UCL

Bio

George Danezis is a Reader in Security and Privacy Engineering at the Department of Computer Science of University College London, and Head of the Information Security Research Group. He has been working on anonymous communications, privacy enhancing technologies (PET), and traffic analysis since 2000. He has previously been a researcher for Microsoft Research, Cambridge; a visiting fellow at K.U.Leuven (Belgium); and a research associate at the University of Cambridge (UK), where he also completed his doctoral dissertation under the supervision of Prof. R.J. Anderson.

Research

His theoretical contributions to the Privacy Technologies field include the established information theoretic and other probabilistic metrics for anonymity and pioneering the study of statistical attacks against anonymity systems. On the practical side he is one of the lead designers of the anonymous mail system Mixminion, as well as Minx, Sphinx, Drac and Hornet; he has worked on the traffic analysis of deployed protocols such as Tor.

His current research interests focus around secure communications, high-integirty systems to support privacy, smart grid privacy, peer-to-peer and social network security, as well as the application of machine learning techniques to security problems. He has published over 70 peer-reviewed scientific papers on these topics in international conferences and journals.

He was the co-program chair of ACM Computer and Communications Security Conference in 2011 and 2012, IFCA Financial Cryptography and Data Security in 2011, the Privacy Enhancing Technologies Workshop in 2005 and 2006. He sits on the PET Symposium board and ACM CCS Steering committee and he regularly serves in program committees of leading conferences in the field of privacy and security. He is a fellow of the British Computing Society since 2014.

Web page: http://www0.cs.ucl.ac.uk/staff/G.Danezis/
Full CV: http://www0.cs.ucl.ac.uk/staff/G.Danezis/danezis-cv.pdf

Warwick

Dr Weisi Guo

Warwick

Bio

Dr. Weisi Guo graduated from the University of Cambridge with degrees in General Engineering and Computer Science. He has worked for 2 years in the telecommunications industry and has been a faculty staff at the University of Warwick for 5 years. In his private life, his experiences include working in UNHCR refugee camps, solo climbing high peaks, and have completed both the London and Sahara full marathons.

Research

His research expertise is in Information and Networks. In terms of applications, he is particularly interested in interdependent critical infrastructures and urban analytics for defence and security. His pioneering research in molecular signalling and complex networks has won him a number of awards & nominations (IEEE, IET, and Bell Labs). At the Turing Institute, he is a Group Leader in the Program on Data-Centric Engineering.

Professor Graham Cormode (ULD, Warwick)

Warwick

Bio

Graham Cormode is a Professor in Computer Science at the University of Warwick and The Alan Turing Institute’s University Liaison Director for Warwick. He works on research topics in data management, privacy and big data analysis. Previously, he was a principal member of technical staff at AT&T Labs-Research in the USA.  He serves as an associate editor for ACM Transactions on Database Systems (TODS), and the Journal of Discrete Algorithms.

Research

Graham will work on a variety of topics at the Turing including:

  • Algorithms for scalable analytics & streaming, sketching, dimensionality reduction, compressed sensing, with applications to internet scale data, vehicle data.
  • Data Anonymization and privacy & statistical and cryptographic approaches to privacy, primarily differential privacy, with applications to telecommunications and social data.
  • Distributed algorithms & algorithms for large scale monitoring and logging of activities, drawing on ideas from approximation and distributed ledger technologies.

Professor Saul Jacka

Warwick

BIO

After a PhD from Cambridge Statistical Laboratory, Saul trained as an actuary and retains strong links with the profession. He has been a Professor of Statistics at Warwick since 2003.
He won two Merrill Lynch Best Paper prizes in Mathematical Finance, is a member of Scientific Steering Committee of the Isaac Newton Institute and editor of the probability journal, Stochastics

 

RESEARCH

Saul specialises in probability, stochastic control and finance.

He is currently working on stochastic control, optimal stopping and applications to finance, anomaly detection and numerics for non-linear pdes and multi-level Monte Carlo (MLMC) simulation.
He is increasingly focusing on high-dimensional and analytically intractable control problems via the Policy Improvement Algorithm, approximations thereof and related numerical tools.

Dr Anthony Lee

Warwick

BIO

Anthony Lee has been an Assistant Professor of Statistics at the University of Warwick since 2013. He received BSc. and MSc. degrees in Computer Science from the University of British Columbia, and a DPhil. in Statistics from the University of Oxford in 2011. He was a Centre for Research in Statistical Methodology Research Fellow from 2011 to 2013.

 

RESEARCH

His research is mainly in the area of stochastic algorithms for approximating intractable quantities that arise in data analysis. Examples of such algorithms are Markov chain and Sequential Monte Carlo. He works on both theory and methodology, with a focus on algorithms that scale well in parallel and distributed computing environments. Research in this area is interdisciplinary, bringing together advances in applied probability, algorithms, architecture and statistics. He is also interested in computational and statistical trade-offs when conducting inference.

Professor Chenlei Leng

Warwick

Bio

Chenlei obtained his BSc in mathematics from the University of Science and Technology of China and his PhD in statistics from the University of Wisconsin-Madison. He held regular and visiting faculty positions in China, Germany and Singapore before joining the University of Warwick as a professor of statistics in 2013.

Research

Chenlei is a statistician working mainly on developing novel statistical methods for analysing complex data. His recent research interests have been focused on high-dimensional data analysis, correlated data analysis, network data analysis and statistical learning. His works have found applications in medicine, biology, engineering, social sciences and so on. As a faculty fellow, he looks forward to the opportunity to work on projects where statistical analysis can be useful.

Dr Filip Rindler

Warwick

BIO

Filip Rindler is currently Associate Professor in Mathematics at the University of Warwick, where he has been since 2013. After completing his doctorate in nonlinear PDE theory and the calculus of variations at the OxPDE centre within the University of Oxford in 2011, he moved to the University of Cambridge to take up the Gonville & Caius College Drosier Research Fellowship, holding the position until 2015 (on leave 2013-2015). For 2014-2017 he is funded by a full-time EPSRC Research Fellowship. Originally a computer scientiest, he will aim to contribute to the ATI’s activities at the interface between theory and implementation.

 

RESEARCH

He is interested in variational / optimization problems, interpreted broadly. He is particularly interested in theoretical aspects, but also in applications to material science, physics, and economics. One particular aspect of his research to date has centered around generalized solutions to minimization problems that do not have a minimizer, but whose approximate solutions develop finer and finer oscillations or concentrations as one approaches the minimum. Describing and analyzing such situations often relies on advanced tools, such as Young measures, and he has contributed several results and techniques in this area.
Recently, he has also been interested in anomaly detection problems.

Dr Krys Latuszynski

Warwick

Bio

Krys Latuszynski is a Royal Society University Research Fellow and Jeff Harrison Early Career Professor of Statistics at the University of Warwick. Prior to this, he was a Research Fellow at the Universities of Warwick and Toronto. He graduated from the University of Warsaw (BSc, MSc, PhD in mathematics) and Warsaw School of Economics (BSc, MSc in econometrics).

Research

His research is in the area of computational statistics with particular focus on design and analysis of Markov chain Monte Carlo and related algorithms for high dimensional problems and unbiased inference for intractable likelihood problems, in particular diffusion process.

Dr Matthew Leeke

Warwick

BIO

Dr Leeke is an Associate Professor and Director of Admissions and Recruitment within the Warwick University Department of Computer Science.

 

RESEARCH

His research addresses a variety of issues relating to the design, implementation and evaluation of dependable systems. He is particularly interested in approaches for the design and location of efficient error detection mechanisms, as well a privacy issues in emerging areas such as wireless sensor networks.

Dr Sebastian Vollmer

Warwick

BIO

He obtained his MSc in Mathematics from the University of Warwick in 2010.  He remained at Warwick to complete his PhD with Profs Andrew Stuart and Martin Hairer, finishing in 2013.  Since October, he is a postdoc working with Profs Arnaud Doucet and Yee-Whye Teh.

 

RESEARCH

His research interests lie at the interface of probability theory and applied mathematics. In particular, he is focusing on the convergence of stochastic processes to their equilibrium. His main area of research is sampling algorithms where the aim is to draw approximate samples from a distribution in order to approximate an expected value. With these methods it is possible to make inference in the areas such as Bayesian statistics, Bayesian inverse problems, molecular dynamics and statistical mechanics. Furthermore, he is interested in dynamical systems and decision theory and he is curious in the handling and use of large data sets.