Computer Science 1.04
If you’ve been wondering what happens at the Turing Institute, what research is being conducted, what ways you could get involved in addition to the Turing Fellow posts, then this day will help to provide some of the answers.
Presentations from Research Fellows, Turing Faculty Fellows and Enrichment year students, as well as discussions on the additional funding streams and opportunities that exist will be given. The University of Warwick Liaison Director, Prof Graham Cormode will also be present to answer questions.
The day will be open for all interested staff, research students and final year students who may be considering continuing their education in data science.
To register your atendance, please complete the booking form.
Suzy Moat - Turing Faculty Fellow
Title: Understanding human behaviour and wellbeing with online photographs
Abstract: Our global society is now uploading increasing volumes of photographs to the Internet, creating vast quantities of data on where we are and the environment we are spending time in. In this talk, I will describe recent work in which we investigate whether online photographs can give us new quantitative insight into human behaviour and wellbeing. I will outline a number of case studies, drawing on deep learning methods and more.
Nathanael Fijalkow - Warwick-Turing Research Fellow
Title: Logic at the Turing
Abstract: When I joined the Turing in January 2017, my background was in automata, logic, games and verification, with a rather theoretical approach.
|12.30||Lunch / poster session from Turing PhD students|
Title: Learning from signatures
Abstract: The theory of rough paths was developed in order to make sense of differential equations driven by paths of arbitrary "roughness". As a byproduct, it created the concept of the signature, which is a different way to present data, sensitive in capturing information on the sequence of events. I will discuss how to use the signature to learn about a system in different types of context and applications.
Adria Gascon - Warwick-Turing Research Fellow
Title: Privacy-preserving data analytics
Abstract: Huge amounts of data exist about every one of us, the use of which has the potential to improve our lives and the world we live in. However, concerns about the privacy of this data have naturally become an increasingly prevalent issue. The aim of privacy-preserving analysis is to utilise this data to its fullest potential without compromising our privacy.In this talk I'll discuss informally some of the basic techniques and challenges in privacy-preserving data analysis, as well as research going on at the Turing in this space.As part of an illustrative example, we'll learn about the relation between Tinder and cryptographic multi-party computation techniques for privacy-preserving computation. I’ll also discuss our results on less romantic - but still privacy-preserving - protocols for private machine learning.
Daphne Ezer - Warwick-Turing Research Fellow
Title: Plant signal integration as a case study of computer-aided experimental design
Abstract: It is important to understand how plants sense and respond to temperature so we can engineer crops to withstand the detrimental effects of climate change. However, it is becoming clear that plants respond to temperature in a context-dependent manner—their response is conditional on other environmental signals such as light and time of day. To understand the mechanism that allows plants to respond to temperature in a context-dependent manner, we need to dissect the structure of the regulatory network that integrates environmental signals. This requires new innovations in computer-aided experimental design, such as the development of new tools to help generate hypotheses, control for biases in existing datasets, and prioritise further experiments to maximise information gained under cost and time constraints. In this presentation, I will use my previous research in plant signal integration as a case study to describe broader questions in computer-aided experimental design that I hope to address during my fellowship at the Alan Turing Institute and at the University of Warwick.
Sebastian Vollmer - Turing Faculty Fellow
Title: Measuring Sample Quality with Diffusions
Abstract: Standard Markov chain Monte Carlo diagnostics, like effective sample size, are ineffective for biased sampling procedures that sacrifice asymptotic correctness for computational speed. Recent work addresses this issue for a class of strongly log-concave target distributions by constructing a computable discrepancy measure based on Stein's method that provably determines convergence to the target. We generalize this approach to cover any target with a fast-coupling Ito diffusion by bounding the derivatives of Stein equation solutions in terms of Markov process coupling times. As example applications, we develop computable and convergence-determining diffusion Stein discrepancies for log-concave, heavy-tailed, and multimodal targets and use these quality measures to select the hyperparameters of biased samplers, compare random and deterministic quadrature rules, and quantify bias-variance tradeoffs in approximate Markov chain Monte Carlo. Our explicit multivariate Stein factor bounds may be of independent interest.
In addition: Some updates on translation projects that came out of the Defence and Security Group with national grid and health projects.
The day will close at 4.30/5.00.
Updates on data science at Warwick can be found at WDSI - https://warwick.ac.uk/wdsi
If you require any assistance, please contact Clare Roberts, DCS - c dot roberts dot 2 at warwick dot ac dot uk