Research

April 28th, seminar by Mel Woods: Future Cities: Co-creating Future City Design Fictions in the Wild

The School of Computer Science welcomes Mel Woods from Duncan of Jordanstone College of Art and Design, University of Dundee.

Abstract:

mel_woodsBlue heritage plaques pepper the UK landscape expounding officially validated narratives celebrating past events, people, and buildings. This seminar will discuss a novel method that draws on this specific cultural context to generate reflective, nano-stories, documenting them through populating a place, physical space, and an online data repository. The guerrilla blue plaque method was designed to support people to reflect on possible futures, in this instance the theme of future cities. The seminar will demonstrate how using critical design artefacts can help support understanding of future hopes, needs, and goals for individuals and communities. It will also discuss the method as a feedback mechanism for participatory design, citizen engagement and emergent outcomes from the latest deployment.

This work was initially developed as part of a UK arts and digital media festival and exhibited recently at Microsoft Research Lab, Cambridge at RTD 2015.

Bio:

Mel is Reader at Duncan of Jordanstone College of Art and Design, University of Dundee. In her research she has developed and explored interaction between people to support discovery, foster creativity and affect. Throughout her academic career she has sustained a critical enquiry in art and design, creating digital artefacts, interfaces, prototypes and exhibits using novel methods and evaluation techniques.

This seminar is part of our ongoing series from researchers in HCI. See here for our current schedule.

Event details

  • When: 28th April 2015 14:00 - 15:00
  • Where: Cole 1.33a
  • Format: Seminar

April 13th, seminar by Nicolai Marquardt: Towards Ad-hoc Collaboration Spaces with Cross-Device Interaction Techniques

Speaker: Nicolai Marquardt, University College London
Date/Time: 1-2pm April 13, 2015
Location: CS1.33a, University of St Andrews

Abstract:
Despite the ongoing proliferation of devices and form-factors such as tablets and electronic whiteboards, technology often hinders (rather than helps) informal small-group interactions. Whereas natural human conversation is fluid and dynamic, discussions that rely on digital content—slides, documents, clippings—often remain hindered due to the awkwardness of manipulating, sharing, and displaying information on and across multiple devices. Addressing these shortcomings, in this talk I present our research towards fluid, ad-hoc, minimally disruptive techniques for co-located collaboration by leveraging the proxemics of people as well as the proxemics of devices. In particular, I will demonstrate a number of cross-device interaction techniques—situated within the research theme of proxemic interactions—that support nuanced gradations of sharing. I will also introduce different novel hybrid sensing approaches enabling these interaction techniques and discuss future research directions.

Bio:
Nicolai Marquardt is Lecturer in Physical Computing at University College London. At the UCL Interaction Centre he is working in the research areas of ubiquitous computing, physical user interfaces, proxemic interactions, and interactive surfaces. He is co-author of the books Proxemic Interactions: From Theory to Practice (Morgan & Claypool 2015) and Sketching User Experiences: The Workbook (Elsevier, Morgan Kaufmann 2012).

This seminar is part of our ongoing series from researchers in HCI. See here for our current schedule.

Following on from Nicolai Marquardt’s successful talk his slides can now be viewed here: St Andrews guest lecture Nicolai Marquardt – Slide Presentation

Event details

  • When: 13th April 2015 13:00 - 14:00
  • Where: Cole 1.33a
  • Format: Seminar

New EPSRC project C3: Scalable & Verified Shared Memory

C3: Scalable & Verified Shared Memory via Consistency-directed Cache Coherence

Dr Susmit Sarkar

Dr Susmit Sarkar

Susmit Sarkar with colleagues in the University of Edinburgh and Intel as project partners, have been successful in their application to the EPSRC for their project C3: Scalable & Verified Shared Memory via Consistency-directed Cache Coherence. This 3 year project starts in July 2015 and aims to realise scalable and verified shared memory.

Shared-memory multi-core processors are ubiquitous, but programming them remains challenging. The programming model exposed by such multi-core processors depends crucially on a “memory consistency model” (MCM), a contract between the hardware and the programmer, which essentially specifies what value a read can return. On the hardware side, one key mechanism to implement the memory consistency model is the “cache-coherence protocol” (CCP), which essentially communicates memory operations between processors. However, the connection between the CCP and the MCM remains unclear. This is especially true for modern CCPs and MCMs, in which CCP design has been divorced from the requirements of the MCM. Susmit and his colleagues argue that this has negatively impacted the scalability and the verifiability of CCPs.

On the scalability front, there are serious question marks about sustaining cache coherence as the number of cores continue to scale. On the verification front, the application of existing verification techniques, which do not verify the CCP against the MCM, are arguably broken.

The C3 proposal, proposes a family of CCPs that are “aware” of, and verified against the MCM. Their approach is motivated by the fact that both hardware and programming languages are converging to various relaxed MCMs for performance oriented reasons. The team use such relaxed MCMs as inspiration to research CCPs that can take advantage of them.

Specifically, they will research “lazy” CCPs where memory operations are batched, and the cost of communicating a memory operation can be amortised. They will also, for the first time, formally verify the relationship between the hardware CCPs and the programmer-oriented MCM they provide. They will investigate rigorously the gains to be had from such lazy CCPs. The team will do this by creating a multi-core silicon prototype of our proposed CCP, leveraging our experience in the design of industrial-strength micro-architectures and their implementations.

March 10th, seminar by Nick Taylor: Sustaining Civic Engagement in Communities

Speaker: Nick Taylor, University of Dundee
Date/Time: 2-3pm March 10, 2015
Location: CS1.33a, University of St Andrews

Abstract:
Engagement with local issues is typically very low, despite digital technologies opening up more channels for citizens to access information and get involved than ever before. This talk will present research around the use of simple physical interfaces in public spaces to lower barriers to participation and engage a wider audience in local issues. It will also explore the potential for moving beyond top-down interventions to support sustainable grassroots innovation, in which citizens can develop their own solutions to local issues.

Bio:
Nick Taylor is a Lecturer and Dundee Fellow in the Duncan of Jordanstone College of Art and Design at the University of Dundee. His research interests involve the use of novel technologies in social contexts, particularly in communities and public spaces. This has involved the exploration of technologies to support civic engagement in local democracy, public displays supporting community awareness and heritage, as well as methods of engaging communities in design.

This seminar is part of our ongoing series from researchers in HCI. See here for our current schedule.

Event details

  • When: 10th March 2015 14:00 - 15:00
  • Where: Cole 1.33a
  • Format: Seminar

Administrative Data Research Centre – Scotland, St Andrews team

The Scottish ADRC is led by Chris Dibben at the University of Edinburgh, and is supported by the Economic and Social Research Council. The ADRC – Scotland:

  • Brings together major Scottish centres of research, and builds on predecessor organisation structures, involving secondary analysis of public-sector data in order to create a common framework for research based on an integrated data linkage service These groups, funded by research councils, charities and Government, include the Scottish Longitudinal Study (SLS), Administrative Data Liaison Service (ADLS), the UK Census and Administrative data LongitudinaL hub (CALLS), Applied Quantitative Methods Network (AQMeN), Scottish Health Informatics Centre (SHIP) and eHealth Research Centre (eHIRC), the national digital data centre (EDINA), Centre for Research on Environment Society and Health (CRESH) and the Centre for Cognitive Ageing & Cognitive Epidemiology.
  • Involves world leading experts in the theory, methods and policy of record linkage for secondary uses, including public engagement, ethics, information governance and law; linkage and analysis of large datasets; geocoding, natural language processing and machine learning. This includes experts from a range of sectors from which administrative data is derived including housing, transport, income, labour markets, health, crime and criminal justice, education, social services.
  • Builds on existing services – enabling immediate access to state-of-the-art facilities for research access to de-identified administrative data by accredited researchers.
  • Co-locates with the Scotland hub of national health informatics research endeavour, which has already brought together key infrastructures, technologists and research groups, enabling synergies and collaborations that will ensure rapid progress towards a national informatics centre of world importance.
  • Exploits Scotland’s unique holding of linked, machine readable, historical administrative data, including the 1932 and 1947 Scottish Mental Surveys, civil registration data (1855-present), Aberdeen Children of the Nineteen Fifties (ACONF), and others, to make available powerful administrative data based cohort and longitudinal studies.
  • Aims to support National Records of Scotland (NRS) in their work exploring alternatives to the traditional decennial based census.
  • Will have a significant programme of public engagement – including working with citizens to produce statistics of use and relevance to them, and press engagement to ensure that accurate messages are reported.
  • The ADRC research programme will inform the entire UK Administrative Data Research Network and produce research – both specific to administrative data use and more broadly social science – world-leading in terms of originality, significance and rigour.

The St Andrews team will lead research in data linkage methodology, and are currently investigating the potential to use prefabricated secure rooms within the premises of institutions where researchers require secure access to sensitive data.

The St Andrews team involves:

  • Graham Kirby, Alan Dearle – researchers
  • Darren Lightfoot – project manager
  • 2-year research fellow to be appointed summer 2015

The data linkage methodology research programme also includes Alasdair Gray at Heriot-Watt and Peter Christen at Australian National University.

CoDiMa (CCP in the area of Computational Discrete Mathematics)

Steve Linton and Alexander Konovalov were successful in the application for the EPSRC-funded Collaborative Computational Project called CoDiMa (CCP in the area of Computational Discrete Mathematics): CoDiMa (CCP in the area of Computational Discrete Mathematics)

CoDiMa is centred on two open source software systems: GAP and SAGE which are already widely used for research and teaching in abstract algebra, number theory, cryptography, combinatorics, graph theory, coding theory, optimisation and search, among other areas.

The CCP aims to support the ecosystem of users, extenders and developers of these systems and encourage best practice in their use, and to support the more rapid uptake of new features such as parallel programming support.

The project will run for 5 years starting from March 1st, 2015.

A PhD studentship, in collaboration with MSR (Cambridge)

Reasoning about Racy Programs under Relaxed Consistency

A PhD studentship, in collaboration with MSR (Cambridge)

http://research.microsoft.com/en-us/collaboration/global/apply-europe.aspx>

Each Microsoft scholarship consists of an annual bursary up to a maximum of three years. The amount varies in different countries and may depend on specific arrangement with public research funding agencies. The bursary continues automatically the following years, provided the Scholar meets the requirements of the institution.

Payment is made to the institution. The amount of the scholarship is the maximum amount Microsoft Research pays to the institution. In addition, every Scholar receives a laptop allowance.

During the course of their PhD, Scholars are invited to Microsoft Research in Cambridge for a PhD Summer School that includes a series of talks of academic interest and posters sessions, which provides the Scholars the opportunity to present their work to Microsoft researchers and a number of Cambridge academics.

Some of the Scholars may also be offered—at the sole discretion of Microsoft Research—an internship in one of the Microsoft Research laboratories. Internships involve working on a project alongside and as part of a team of Microsoft researchers. Scholars are paid during their internship—in addition to their scholarship bursary. Interested Scholars can apply through the Microsoft Research internship website.

 

School Seminar Series: Matching in Practice: Junior Doctor Allocation and Kidney Exchange

Matching in Practice: Junior Doctor Allocation and Kidney Exchange by Dr. David Manlove

Abstract:
Matching problems typically involve assigning agents to commodities, possibly on the basis of ordinal preferences or other metrics. These problems have large-scale applications to centralised matching schemes in many countries and contexts. In this talk I will describe the matching problems featuring in two such schemes in the UK that have involved collaborations between the National Health Service and the University of Glasgow. One of these dealt with the allocation of junior doctors to Scottish hospitals (1999-2012), and the other is concerned with finding kidney exchanges among incompatible donor-patient pairs across the UK (2007-date). In each case I will describe the applications, present the underlying algorithmic problems, outline the computational methods for their solution and give an overview of results arising from real data connected with the matching schemes in recent years.

BIO:
David Manlove is a Senior Lecturer at the School of Computing Science, University of Glasgow, where he has been since 1995. His research interests lie mainly in the field of algorithms and complexity, and include algorithms for matching problems involving preferences. These arise in applications such as the assignment of school leavers to universities, kidney patients to donors and junior doctors to hospitals. He and his colleagues have been involved in collaborations with the NHS in relation to the Scottish Foundation Allocation Scheme (for matching junior doctors to hospitals) and the National Living Donor Kidney Sharing Schemes (for enabling kidney “swaps” between incompatible donor-patient pairs) where optimal matching algorithms developed by him and colleagues have been deployed. He has over 50 publications in this area including his book “Algorithmics of Matching Under Preferences”, published in 2013.

Event details

  • When: 3rd February 2015 14:00 - 15:00
  • Where: Cole 1.33a
  • Series: School Seminar Series
  • Format: Talk

School Seminar Series: Statistically Consistent Estimation and Efficient Inference for Natural Language Parsing

Statistically Consistent Estimation and Efficient Inference for
Natural Language ParsingBy Shay Cohen, University of Edinburgh.

Abstract:
In the past few years, there has been an increased interest in the machinel earning community in spectral algorithms for estimating models with latent variables. Examples include algorithms for estimating mixture of Gaussians or for estimating the parameters of a hidden Markov model.

The EM algorithm has been the mainstay for estimation with latent variables, but because it is guaranteed to converge to a local maximum of the likelihood, it is not a consistent estimator. Spectral algorithms, on the other hand, are often shown to be consistent. They are often more computationally efficient than EM.

In this talk, I am interested in presenting two types for spectral algorithms for latent-variable PCFGs, a model widely used in the NLP community for parsing. One algorithm is for consistent estimation of L-PCFGs, and the other is for efficient inference with L-PCFGs (or PCFGs). Both algorithms are based on linear-algebraic formulation of L-PCFGs and PCFGs.

BIO:
Shay Cohen is a Chancellor’s fellow (assistant professor) at the University of Edinburgh (School of Informatics). Before that, he was a postdoctoral research scientist in the Department of Computer Science at Columbia University, and held an NSF/CRA Computing Innovation Fellowship. He received his B.Sc. and M.Sc. from Tel Aviv University in 2000 and 2004, and his Ph.D. from Carnegie Mellon University in 2011. His research interests span a range of topics in natural language processing and machine learning, with a focus on structured prediction. He is especially interested in developing efficient and scalable parsing algorithms as well as learning algorithms for probabilistic grammars.

Event details

  • When: 21st January 2015 14:00 - 15:00
  • Where: Cole 1.33a
  • Series: School Seminar Series
  • Format: Talk