Past Seminars
Seminars held over the last six months.
- 04/11/2009: John Richards - Software Productivity Assessment
Following a quick tour of the Software Engineering Department at IBM's Thomas J. Watson Research Center, I'll review our work in support of the creation of a new, commercially viable, peta scale super computer. The fact that we are crossing the barrier to sustained petaflop performance is interesting. Even more interesting is the fact that the project has aggressive requirements for productivity, not just performance. This reflects the field's growing realization that a major barrier to further advances in parallel and scientific computing is the ability of humans to write, debug, tune, deploy, and administer software at this scale. I'll discuss how we are assessing improvements in human productivity attributable to new parallel programming languages and new tools and environments. I'll conclude with a few thoughts on how productivity assessment can be woven into the more general practice of software engineering.
Bio
John Richards joined the Computer Science research staff at IBM's Thomas J. Watson Research Center in 1978 after receiving his Ph.D. in Cognitive Psychology. He has served in research, programming and management roles in numerous networking and interpersonal communications projects and has been recognized for his contributions in the area of digital voice mail systems by the Human Factors Society. Most recently, he has been involved in the design and programming of solutions in areas including Internet access for students and teachers, wireless e-business, web accessibility, social computing, end-user programming, and software productivity assessment.
Dr. Richards has been active in the ACM SIGCHI, SIGPLAN, and SIGACCESS communities. He chaired the OOPSLA'91 Conference and served as chair of the OOPSLA Steering Committee from 1991 to 1996. He was elected a Fellow of the ACM in 1997 and a member of the IBM Academy of Technology in 2006.
Host: Vicky Hanson - 28/10/2009: Dr Craig Robertson - Adventures in High Performance Machine Vision and Visualization
As a practitioner in the field of machine vision I have been involved in many exciting and demanding projects from 3D reverse-engineering to prototype medical devices. In this talk, I outline some of the work achieved and illustrate just how much scope there is to this rapidly expanding field.
I will develop the emergent theme that advances in hardware have impacted on both the speed and range of computability of complex machine vision algorithms and allowed the recent sudden uptake of machine vision rich products at a consumer level.
Host: Manuel Trucco - 21/10/2009: Mamie Bruce-Gardyne - How I Learned to Love TeleHealth and My Battle to Convince the GPs!
This seminar is at 13:00
Mamie Bruce-Gardyne is the founder of the award-winning Airlie Silver Surfers Cybercafe - still going strong after 10 years (this week they in fact are marking their 6000th attendance). Mamie's latest campaign is to get a TeleHealth initiative off the ground, to benefit older people in the remoter parts of Angus (and not so remote, if they are having mobility problems). She will be showing a professionally produced DVD to promote the ideas which she was involved in making, and will be sharing with us a personal view from the trenches (in her own quite inimitable style!).
A session not to be missed by all of those involved in older people and computing, healthcare computing and HCI.
Host: Norman Alm - 14/10/2009: Rachel Menzies and Scott Loynton - Two Postgraduate Research Presentations
Rachel Menzies - Promoting sharing behaviours in children through the use of a customised novel computer system.
ECHOEs is a research project that aims to develop an adventurous technology-enhanced learning environment in which both typically developing children and children with Asperger Syndrome (a form of autism) can explore and improve social interaction and communications skills. As part of this research project, I am investigating ways in which to promote sharing behaviours in children through the use of this novel computer system. I am interested in investigating the effect that customisation of the environment will have on the effectiveness of the system in improving sharing behaviours. Participatory Design will be employed throughout the project with teachers, therapists, parents and children, both typically developing and with Asperger Syndrome.
Scott Loynton - Academic scientific software development projects aim to provide valuable research tools to aid the scientific community in the process of discovery. The need for those tools to meet wider expectations of usability and good user experience design is critical to success; however the nature of academic e-science software development projects means they are often constrained by too narrow an understanding of the potential user base, and a focus on development at the expense of broad user research. We introduce the concept of the 'Scientific Software Community Framework' as a way of addressing these limitations in capacity to support effective “user research” that informs and influences longer-term marketing and outreach strategy as well as supporting the development of usable and useful scientific software. - 07/10/2009: Shehroz Khan - One Class Classification: Review and Application to Raman Spectroscopy
The One Class Classification (OCC) problem is different from the conventional binary/multi-class classification problem in the sense that in OCC, the negative class is either not present or not properly sampled. The problem of classifying positive (or target) cases in the absence of appropriately-characterized negative cases (or outliers) has gained increasing attention in recent years. Researchers have addressed the task of OCC by using different methodologies in a variety of application domains. In this talk, a taxonomy with three main categories will be presented based on the way OCC has been envisaged, implemented and applied by various researchers in different application domains. We then present an application of OCC for the identification of chemical spectral data using Raman Spectroscopy.
Host: Jesse Hoey - 30/09/2009: Wei Jia - Postgraduate Research Presentation
Segmentation, categorisation and retrieval of fabric design images
This talk will discuss segmentation of digital images and the usage for image categorisation and retrieval in a historical commercial archive owned by Liberty Fabric.
Our research involves building a good model to segment these fabric images and using the segmentation results to represent images for image retrieval and categorisation. The segmentation of fabric design is formulated as a pixel labelling problem. Algorithms based on Markov random field (MRF) optimisation and re-estimation are described. The performance of MRF models optimised using alpha-expansion and iterated conditional modes, both with and without parameter re-estimation are evaluated quantitatively. Based on the segmentation result, images are segmented into separated shapes which are represented by Generic Fourier descriptor. We are seeking feasible and efficient ways to use these local features for image retrieval and categorisation.
- 23/09/2009: Deborah Fels - Inclusive entertainment:
Multimodal interfaces for access to entertainment by people with sensory disabilities
Having access to entertainment and media is as important as access to work and daily living tasks for people with disabilities. We have been working on cross-modal and sensory substitution techniques for access to film, television and theatre. One key thesis that we employ is that film, television, and theatre are important forms of cultural expression and people consume them for the express purpose of entertainment. Any access feature should also consider that perspective but most do not because access is not considered part of the entertainment package but a simple add-on at the very end of the process. I will present our research in enhanced captioning, music visualization, vibrotactile music and audio description along with entertaining examples that demonstrate the concept of inclusive entertainment.
Short bio
Dr. Deborah Fels has a PhD (1994) in Human Factors from Industrial Engineering at the University of Toronto, and a Masters of Health Science (1987) in Clinical Engineering from the University of Toronto. She is currently employed as an associate professor in the School of Information Technology Management, and the Director of the Centre for Learning Technologies at Ryerson University. She is currently on sabbatical at the University of Dundee and working with Dr. Norman Alm.
Her research interests include access to multi-media computer applications for people with disabilities, accessible media, computer-based learning, and video conferencing. Current research projects include: 1) interactive video-mediated communication system for children in hospitals; 2) emotive captioning; 3) broadband learning environments supporting people with disabilities; 4) audio description; 5) SignLink Studio co-creator and on-line ASL see www.signlinkstudio.ca and 6) inclusive game design. She received one of Canada’s Top 40 Under 40 awards for the year 2001. She is also a professional engineer.
- 16/09/2009: Lorna Gibson - Developments to the User Centre and the Usability Lab
We will present to developments that will take place over the next
couple of months to both the User Centre and the Usability Lab. We will cover how the physical space will change, what this will be able to offer and how the use of these spaces will be affected.
- 09/09/2009: Xinyi Jiang - How to improve my news writing skills?
What is news? What makes news news? What do we want to know from news? How to attract readers’ attention with a good news writing style?
Xinyi Jiang, currently develops academic collaborations in the University’s College of Art, Science & Engineering, previously a design ethnographer with us, also spent years studying and teaching subjects including English, journalism and media studies. She volunteers to help the staff and students from the School of Computing to improve our news writing in order to communicate and promote our achievements more effectively and successfully.
Even though you do not want to become a journalist, this workshop will be useful for you to understand some basic principles of news writing and communicating skills, which is no less important to computer scientists than, sociologists or journalists. A similar workshop held on the 3rd this month among a small group in SoC was well-received and attendees said that it was ‘eye-opening’ and ‘gave me a lot to think about’.
- 02/09/2009: Martin Dempster and Joseph Devereux - Two Postgraduate Research Presentations
- 22/07/2009: Prof. Henry Prakken - SICSA Distinguished Visitor - Sense-making software for fact finding in law.
In dealing with evidential problems, lawyers and crime investigators are often faced with a large body of evidence of which they have to make sense. They have to map out the reasonable hypotheses about what might have happened and they must indicate how the available evidence supports or attacks each of these hypotheses. This is a difficult and laborious task where mistakes are easily made. Sense making systems are a computer-based means to support humans in this task. Such systems support humans with graphical means in formulating their hypotheses, in relating them to the available evidence and in expressing their arguments about why certain evidence supports or attacks a certain hypothesis. Ideally, sense-making systems can also indicate the plausibility of a hypothesis in light of the evidence.
In this talk a sense-making system called Avers is discussed, which is being developed at the universities of Utrecht and Groningen. The main focus will be on the underlying reasoning model, which combines defeasible argumentation with abductive scenario construction.
------ Bio ---------------
Henry Prakken is professor in Law and IT at the Law Faculty of the University of Groningen and lecturer in Intelligent Systems at the Computer Science Department of Utrecht University (The Netherlands). His main research interests concern logical and dialogical aspects of argumentation, and the application of argumentation in legal reasoning, multi-agent systems and other domains. With Gerard Vreeswijk he wrote a chapter on 'Logical systems for defeasible argumentation' in the Handbook of Philosophical Logic (2nd edition). Prakken has been a key note speaker at several conferences, including COMMA-2008, the 2nd International Conference on Computational Models of Argument. He is the current President of the International Association for Artificial Intelligence and Law (IAAIL).
Host: Chris Reed - 15/07/2009: Adam Wyner - From Arguments in Natural Language to Argumentation Frameworks
With apologies for the short notice, there is an additional seminar today. Adam Wyner from University College London will be talking on at noon today. We will be in the seminar room, rather than in Wolfson.
Host: Chris Reed - 24/06/2009: Dr Oliver Schreer - Advances in 3D Videocommunication
***Note this Seminar will be running at 12.00-13.30**
Due to the advances in computer vision, hardware development and display technologies, 3D video processing has achieved a lot of attention in the last few years even in the field of 3D videocommunication. Many research activities have been started on European level in order to investigate 3D video processing for different domains such as 3DTV, 3D cinema, 3D on mobile devices and 3D videoconferencing. The aim of this talk is to give an overview of current trends and perspectives in two domains of 3D videocommunication, namely 3D videoconferencing and 3DTV. In this context, latest research results will be presented in different scientifically challenging fields such as multi-view analysis, view rendering, real-time 3D video processing, multiview displays.
Host: Manuel Trucco - 17/06/2009: Karen Petrie - Using Constraint Programming Techniques to Solve Problems - **This Seminar will be at 2pm**
Problems often consist of choices. Constraint programming (CP) is the branch of Artificial Intelligence, where computers help us to make these choices. Constraint programming is a multidisciplinary technology combining computer science, operational research and mathematics. Constraints arise in design & configuration, planning & scheduling, diagnosis & testing, and many other contexts.
A constraint program consists of variables, a set of possible values for each variable, and a number of constraints which must be satisfied. For example, the problem might be to fit components to circuit boards, subject to the constraints that no two components can be overlapping and the circuit board is as small as possible. A solution to a CSP is an assignment of values to variables such that none of the constraints are violated.
CP can solve problems in telecommunication, e-commerce, electronics, bioinformatics, transportation, network management, supply chain management, and many other fields. In this talk I will explain exactly what CP is and how it can be used to solve practical problems.
Host: Peter Gregor - 03/06/2009: Prof. Guillermo Simari - Modeling the Accrual of Arguments in Defeasible Logic Programming
Argumentation frameworks have proven to be a successful approach to formalizing commonsense reasoning. Recently, the notion of accrual of arguments has received attention from the argumentation community. Three principles for argument accrual have been identified as necessary to hold in argumentation frameworks. In this talk we we will propose an approach to model the accrual of arguments in the context of Defeasible Logic Programming, a logic programming approach to argumentation which has been successful for some real-world applications. We will analyze the above mentioned principles in the context of our proposal, studying other interesting properties. Recently, some argumentation frameworks have been extended to deal with possibilistic uncertainty, notably Possibilistic Defeasible Logic Programming (P-DeLP). Here, we will outline a novel approach to model argument accrual in the context of P-DeLP in a constructive way.
Short Vita
Guillermo R. Simari currently holds the position of Chair of the Computer Science and Engineering Insitute, Universidad Nacional del Sur, in Bahia Blanca, Argentina. The focus of his research is on the formal foundations, and effective implementation, of Defeasible Reasoning Systems for Autonomous Agents, having published more than a hundred articles on the topic. This research is carried out as leader of the Artificial Intelligence Research and Development Laboratory
(LIDIA) where several lines of research are pursued, among which are:
Reasoning under Uncertainty, Negotiation, Belief Revision, Computational Logic, Reasoning on the Web. The Lab is also involved in applied research and development in multiagent systems and robotics for a wide range of applications of interest to government and industry. This research has produced DeLP (Defeasible Logic Programming), a fully implemented system which combines Logic Programming and Defeasible Argumentation. G. R. Simari has recently co-chaired a Dagstuhl Perspectives Workshop on the Theory and Practice of Argumentation Systems, a LPNMR Workshop on Argumentation and Non Monotonic Reasoning, and the International Workshop on Argumentation, Dialogue, and Decision Making during the 2006 Non-Monotonic Reasoning meeting in England. He has served as reviewer and PC member in International Conferences and Workshops such as AAMAS, IJCAI, AAAI, ECAI, COMMA, NMR, ArgMAS, CLIMA, and others. He has also given a Tutorial on Defeasible Logic Programming in ICLP 2004, a course on Computational Models for Argumentation in MAS as part of EASSS 2005, and more than 25 invited talks worldwide. He is coeditor of the new Taylor & Francis Journal Argument and Computation together with C.
Reed, F. Grasso and I. Rahwan. He has edited together with I. Rahwan the book Argumentation in Artificial Intelligence that will appear in August 2009.
Host: Chris Reed - 20/05/2009: Simone O’Callaghan - Research Presentation
Graphical Tagging as an Emerging Art Practice in Mobile Media
Simone O’Callaghan is an art-practice based researcher, examining the ways in which mobile technologies are being combined with art practices to explore relationships between physical, virtual and augmented spaces. She will discuss her research into the use of graphical tagging in contemporary art and media, what it is (2-dimensional barcodes, such as QR-codes, readable by mobile phone), and the current issues from a global perspective surrounding mass market adoption of such technologies.
Using examples of her own work from recent exhibitions she will discuss the challenges she has encountered making artworks in an area where the lack of technical standards and device fragmentation are obstacles which deter many artists and developers. She will also explore ways innovative uses of graphical tagging and what this may mean for future creative practices using mobile devices.
Any questions, comments or speaker suggestions? Contact Gemma Webster.