Postgraduate Prospectus 2008

Computing, artificial intelligence and IT research information

Research groups

Research is organised around well-defined groups of international standing. Some of our research is highly interdisciplinary, involving collaborations between these groups as well as with other departments at Sussex. All groups are well funded from a variety of sources, including research council grants and support from industry, and all have specialist facilities and laboratories. The research groups are briefly described below.

Centre for Research in Cognitive Science (COGS)

This is an interdisciplinary Research Centre, which aims to champion and support research and teaching in cognitive science at Sussex.

A world-renowned, pioneering institution in cognitive science, COGS continues to conduct first-class research in such topics as computer vision, mental representation, cognitive linguistics, conscious experience, computational architectures for emotion, machine learning and neural networks, pattern recognition, cognitive modelling, implicit learning, reasoning, mechanisms of creativity, philosophical issues in artificial life, and temporal cognition in language.

The Centre acts as a focus for research in cognitive science by running interdepartmental seminars and research meetings. For more information, contact the Centre’s director, Ron Chrisley (r.l.chrisley@sussex.ac.uk).

Centre for Computational Neuroscience and Robotics (CCNR)

This thriving interdisciplinary group is jointly run by the Department of Informatics and the Neuroscience group in the School of Life Sciences. It has its own well-resourced laboratories and focuses on work at the interface between computing structures and biological systems – an area that has been recognised by government funding agencies and the EU as an emerging area of considerable importance. The symbiosis between computer science and neuroscience in particular holds the key to future developments in robotics and artificial intelligence. It will lead to a better understanding of how the brain works and promises biomedical advances of enormous benefit.

Artificial intelligence and neuroscience are areas in which Sussex is exceptionally strong. Combined with a tradition for interdisciplinary links between the School of Science and Technology and the neuroscience groups (IRC, EP and MRC laboratory) in the School of Life Sciences, this places Sussex at the forefront of this emerging field.

Work in the Centre mainly falls under the following headings: modelling neural systems, evolutionary and adaptive robotics, evolutionary electronics, insect and robot navigation, theory of natural and artificial evolution and computational creativity. This last area uses adaptive technology in the creative arts and involves various collaborations with local and international artists and performers.

The CCNR overlaps with the Evolutionary and Adaptive Systems group, described below. For more information, contact the directors: Professor Phil Husbands (p.husbands@sussex.ac.uk) or Professor Michael O’Shea (m.o-shea@sussex.ac.uk).

Computer vision and medical imaging lab

This group uses a wide variety of methods and approaches from computer science, cognitive and biological sciences to research into a variety of areas including: dynamic aspects of visual perception, low level vision, spiking neural nets, data mining and machine learning.

Faculty involved include: David Young, Des Watson, Professor Phil Husbands, and Andy Philippides.

Creative Systems Lab

The research of this group lies at the intersection of arts, science and technology. Members of the group work on theories of the creative process and their implementation in computer models and artistic productions; they also initiate and facilitate collaborations between scientists, artists, musicians, industry and the general public. In addition to those listed below, faculty involved include: Professor Margaret Boden, Ron Chrisley, Professor Phil Husbands, and Andy Philippides.

Faculty research interests include:

Nick Collins Live electronic music and audiovisuals; interactive music systems; music understanding by computer; algorithmic composition and sound synthesis; mathematics and psychology of music.

Chris Thornton Computational learning using symbolic algorithms and connectionist mechanisms; theories of creativity.

Evolutionary and adaptive systems

This group is concerned with the interface between the biological and computational sciences and applications of new technologies resulting from work in this area. Research focuses on a wide range of topics including: the development of biologically inspired adaptive algorithms, adaptive and evolutionary robotics, artificial life, computational biology, computational neuroscience, dynamical systems approaches to cognition and development, evolutionary electronics and evolutionary computation, evolutionary theory, and applications of adaptive systems in the creative arts.

There is also a strong line of work in bioinformatics and machine learning. In machine learning, research involves the development of novel and efficient algorithms, building on statistical theory as well as borrowing ideas from neuroscience. Research in bioinformatics includes microarray data processing, MD and protein structures.

This well-resourced group overlaps with CCNR and, together, they make up one of the largest and best-known research groups in the world working in this area. The group collaborates widely with other centres in the University and elsewhere, has a number of well-equipped laboratories, machine shop facilities and runs several lively seminar series, as well as smaller specialist discussion groups.

Faculty research interests include:

Luc Berthouze Motor development in infants and in machines; clinical applications of dynamical system approach to characterising infant movements (particularly cerebral palsy); EEG-based brain-machine interfaces, intelligent neuro-prostheses; epigenetic or developmental robotics, modelling cognitive development with robotic systems.

Ezequiel Di Paolo Evolutionary robotics, evolutionary biology and embodied cognitive science; computational models of the evolution of social behaviour, altruism and coordination through acoustic interactions.

Inman Harvey Artificial evolution as applied to design: theoretical (error thresholds, neutral networks, optimising speed of search) and applications (evolutionary robotics, evolvable hardware, combinatorial chemistry).

Professor Phil Husbands Evolutionary and adaptive robotics; evolutionary computation and optimisation; artificial life; computational neuroscience; adaptive systems; neuromodulation; history of AI and cybernetics; creative systems.

Thomas Novotny Use of dynamical systems theory, statistics and hybrid systems experiments in a comprehensive approach to understand information processing in nervous systems; information processing in olfactory systems and applications to electronic noses; sequence learning in neuronal systems; accurate conductance-based neuron models and hybrid systems.

Andy Philippides Computational neuroscience and neuroetholgy; evolutionary robotics; insect visual homing strategies; gaseous neuromodulators in real and artificial neural networks.

Anil Seth Theoretical neuroscience and evolutionary and adaptive systems; time-series analysis of neural dynamics, causality in neural systems, neurorobotics, neural mechanisms of consciousness, and network-theoretic approaches to the analysis of complex systems; evolutionary theory and ecological modelling.

Adrian Thompson The application of artificial evolution to engineering design, principally electronics; custom/reconfigurable computing; fault-tolerance; evolutionary theory.

Si Wu Computational neuroscience, machine learning, bioinformatics and neural coding.

David Young Computational and biological aspects of vision, including theoretical and experimental research on optic flow and image representation, and practical applications such as traffic monitoring.

Foundations of computation

This group focuses on the foundational aspects of computer science, particularly the semantics of computation. They have developed behavioural theories for a range of process languages, including features such as higher-order abstractions and distributed resources.

Research interests in this thriving group include:

Professor Matthew Hennessy Semantic theories for programming and specification languages; foundation of concurrency; development and implementation of verification tools.

Bernhard Reus Mathematical semantics of programming languages; their foundations, ie Domain Theory and Type Theory; Synthetic Domain Theory as the natural synthesis of both; constructive (categorical) logic; programming logics; formal proof; formal techniques and tools supporting (object-oriented) program analysis, design and verification.

Human-centred technology (HCT)

The HCT group is an internationally renowned research centre in HCI, interaction design and interactive learning environments. Its focus is on understanding how people interact with and communicate through technology, with particular interests in more cutting-edge tangible and pervasive technologies.

The group is made up of two research labs: the INTERACT lab and the IDEAS lab. One of the main strengths of the HCT group is its interdisciplinary heritage, with faculty, research fellows and DPhil students coming from a variety of backgrounds, including computer science, psychology, artificial intelligence, engineering, art, design and philosophy.

A common theme to all HCT research is putting people at the centre of the design process. Key strands of research include: interactive learning environments and technology-enhanced learning, tangible embodied interaction, pervasive technologies in the home and nonwork environments, pervasive care/telehealth, psychology of programming, accessibility and game design.

Faculty research interests include:

Professor Ben du Boulay Application of AI techniques to education, particularly modelling motivation and metacognition in intelligent learning environments.

Geraldine Fitzpatrick People-centred design methods and conceptual frameworks, especially for emerging technologies such as mobile, tangible and ubiquitous computing and their applications to everyday contexts; special interests in social interaction and collaboration, creativity/play, domestic environments, older people, disability, and healthcare/telecare.

Judith Good Constructivist learning environments; educational games and simulations; technology toolkits for learning; visual programming languages; embodied and tangible interfaces for learning.

Pablo Romero Exploring the support that collaboration and external representations can offer to students learning programming and the potential that programming and software design have for skill transfer into other areas of learning; exploring the potential that new forms of interaction (tangible interfaces, embodied interaction) have for learning in general and specifically as a way to approach engagement and motivation; as a way to foster collaboration; and for specifying concrete and abstract behaviours (programming, scripting) and the impact that this has on learning.

Blay Whitby Social and ethical implications of AI and ALife; philosophical foundations of AI and ALife; professionalism in computing; multimedia and decision support systems.

Natural language processing

This group comprises one of the largest teams of researchers in the UK focusing on statistical and corpus-based approaches to natural language processing. Their research covers probabilistic processing techniques, linguistic theory and automatic acquisition of grammatical knowledge from corpus data, with application to practical natural language parsing and generation.

The group currently consists of about 20 faculty, doctoral and postdoctoral researchers. It has a long and distinguished track record at international level, with many achievements in basic and applied research. Current work covers six related areas of research: probabilistic and robust parsing, annotation of text and transcribed speech, automatic acquisition of lexical information from corpora, computational formalisms for representing information about language, empirical foundations of language processing and practical applications of language processing.

Faculty research interests include:

Professor John Carroll Hybrid linguistic/ statistical approaches to disambiguation of text, efficient parsing, evaluation of parser accuracy, tools for large-scale natural language grammar and lexicon development, and linguistic approaches to the generation of text from representations of its meaning.

Bill Keller Applying machine learning techniques to problems in language learning/grammatical inference.

Rudi Lutz Machine learning, especially of language models (eg grammars, or Hidden Markov Models) using various techniques (ie evolutionary, expectation maximisation).

Professor Geoffrey Sampson Corpus-based natural language processing using statistical stochastic optimisation techniques; standards definition for natural language computing.

David Weir Controlling non-determinism in natural language generation, using language in pervasive computing environments, efficient parsing with large grammars, probabilistic parse ranking, and inferring knowledge about words from raw text.

Philosophy of artificial intelligence and cognitive science

This group considers the conceptual issues that arise when trying to explain natural intelligence systems, or create artificial ones.

Faculty research interests include:

Professor Margaret Boden Computational approaches in the philosophy of mind and theoretical psychology: special interest in purpose, intention, motivation and creativity; philosophy of AI and ALife; social implications of AI.

Ron Chrisley Non-conceptual representations and human cognition; computational architectures for psychological theories of preobjective mind, consciousness, philosophy of computation, computational architectures for affect and emotion.

Other faculty, who have primary affiliations within disciplines such as linguistics, philosophy, neuroscience and psychology, play an active role in this group.

Representation and cognition

Group members are interested in the development and application of cognitive theories. They study the higher forms of cognition including reasoning, problem solving and learning.

The group has a particular focus on the role(s) of representation, emphasising cognitive and semantic dimensions. Favoured methodologies include rich data capture, protocol analysis, experimental designs, the use and design of technology to test theory, and modelling.

Faculty research interests include:

Professor Peter Cheng The nature of representational systems, both external in the world and internal to the mind. A particular interest is diagrams that support advanced forms of cognition, such as complex problem solving, discovery and conceptual learning.

Richard Cox Human reasoning, especially with external representations such as diagrams; representational systems; vicarious learning; artificial intelligence and educational systems; interactive learning environments.

Sharon Wood Multi-agent systems, especially anticipating agent behaviour through situational cuing and intention recognition. Situational modelling, in particular, the acquisition of information through cognitively plausible visual attention processes, and the acquisition of situational understanding through cognitively plausible epigenetic processes, cognitive and epigenetic robotics.

Software systems

Staff and postgraduate students within this group have research interests covering a wide range of topics including programming language design and implementation, object-oriented programming and design, software development environments, debugging tools, networking, communication, operating system design, compiler design, parallelism, code optimisation, automated code generator construction and software development for embedded systems. Many of the current projects have commercial involvement, and we are keen to develop further links with industry.

Particular research strands include the development of compiler technology for embedded systems, and user-centred networking, where the needs and constraints of the users drive the engineering of networked systems. These two areas are coming together in the development of programming languages for pervasive computing.

This well-funded group has its own specialist facilities and laboratories.

Dan Chalmers Context awareness and the way in which system behaviour impacts the user experience in ubiquitous/pervasive computing scenarios; how knowledge of context (including resource limits, location, and other physical and social aspects of context) can be used to modify behaviour, affect data display and configuration of systems.

Ian Wakeman Communications and distributed systems, distributed multimedia and collaborative working.

Des Watson High-level language compilers; code generator design and implementation; sensor networks; medical computing, particularly computer support for nuclear magnetic resonance imaging and spectroscopy.

Space science

The University of Sussex Space Science Centre is an interdisciplinary cross-departmental research centre based in the School of Science and Technology.

The Space Science Centre is led by Professor Paul Gough and conducts both general and mission-oriented space research in close collaboration with other space research institutes in Europe, the US, and the former Soviet Bloc countries. The group designs and constructs instruments, and continues to monitor their operation once launched into space with subsequent scientific data analysis.

The group has probably placed more computers in space than any other UK university space group. It has led with on-board intelligence within space instruments, and claims the first geophysical phenomenon to be discovered by a neural network ‘Expert Data Analyst’ flown within a space instrument.

The Space Science Centre attracts a lot of external funding and the research interests of the faculty involved, Professor Paul Gough, Natalia Beloff and Andrew Buckley (Department of Engineering and Design), include the following areas: space instrumentation, space plasma diagnostics and scientific interpretation, particle correlation technique, intelligent instruments, smart autonomous instruments, real-time data analysis in space instruments, embedded systems, data compression, parallel processing and faulttolerance versus artificial neural networks for data classification and analysis, associative list memory for data classification, fuzzy logic in control of instruments, evolutionary instruments to adapt to unforeseen environments, graphical display and dissemination of complex data sets for rapid man-machine interaction, knowledge accumulation from databases, remote data gathering and processing for unmanned instruments in inhospitable locations, satellite communication systems.

Centre for VLSI and Computer Graphics

The Centre’s research focus has moved from one of developing innovative graphics accelerator technologies at the low hardware and software algorithm level – we designed one of the first graphics accelerators leading towards the birth of the PC graphics card industry – to taking a high-level design and simulation approach that also encompasses the environment as part of the simulation. The Centre has now established a research theme focused on ‘modern living extensions’, based on developing pervasive simulation environments that allow real-world environments, such InQbate – The Centre for Excellence in Teaching and Learning in Creativity, to be simulated, created, technologically enhanced and evaluated. Simulation of pervasive computing environments represents an interesting and challenging research area for our postgraduate students.

The Centre is also engaged in research focused on the design and development of innovative ‘digital heritage systems’, where we explore the use of ICTs including virtual and augmented reality, and interaction technologies to the cultural heritage domain. Here we seek to understand how people could appreciate their heritage through the pioneering design of heritage systems. In this respect the Centre has developed two innovative digital heritage systems: the ARCO Virtual Exhibition System, which is now being licensed and commercially exploited by museums, and the EPOCH Multimodal Interface that allows users to interact with a physical museum artefact through the medium of a virtual replica or simulation environment. This research is underpinned by detailed usability trials and evaluations in actual museums in order to assess the efficacy of these digital heritage systems.

The Centre is also exploring the idea of ‘fidelity’ of simulations in relation to their real world counterparts. A central issue is the investigation of perceptual sensitivity measures of simulation engineering limitations, such as latency and rendering quality. Functional fidelity metrics that are based on spatial cognition aspects such as memory awareness states and schemas have been established and incorporated into perceptually based real-time rendering engines. Through this work, the Centre has established an international reputation working with collaborators such as HP Labs and NASA Ames Research Centre, USA.

Faculty research interests include:

Katerina Mania Fidelity metrics for computer graphics simulations; perceptually based computer graphics rendering; display technologies; 3D user interfaces; human factors issues for virtual environments (task performance after-effects); presence; vection and latency for immersive simulations; visualisation.

Paul Newbury Multimedia systems, in particular virtual prototyping and MPEG2/MPEG4 streaming data. Image processing, image compression, lossy compression of multivariate scientific data sets; analysis and manipulation and compression of deep multispectral electron microscope images.

Adrian Thomas Non-invasive techniques for shape capture to digitise surface structure of faces, feet etc; real-time extensions for gait capture and movement analysis.

Phil Watten Software development; virtual prototyping; high-level design; system modelling; display systems; interface design; and all aspects of media production including new media and web broadcasting.

Martin White 3D graphics; virtual, augmented and mixed reality applied to digital heritage systems; digital libraries; semantic-based knowledge and content systems; technologyenhanced learning; access to and preservation of cultural and scientific resources; virtual archaeology; virtual reconstruction. Metadata standards for digital heritage, interoperable systems; e-learning and information technology; knowledge GRID applied to digital libraries; e-Government, information visualisation applied to independent living and inclusion; and digital content creation.

Research interests

For further information on faculty research interests, refer to the faculty profile pages.

Subject information

Read the core subject information.

For pre-application enquiries:

Student Recruitment Services
Sussex House
University of Sussex
Falmer, Brighton, BN1 9RH
T +44 (0)1273 876787
F +44 (0)1273 876677
E pg.enquiries@sussex.ac.uk

For post-application enquiries:

Postgraduate Admissions
Admissions Office
Sussex House
University of Sussex
Falmer, Brighton, BN1 9RH
T +44 (0)1273 877773
F +44 (0)1273 678545
E pg.applicants@sussex.ac.uk

Teaching term dates 2008-2009

Autumn term
6 October 2008 to 12 December 2008

Spring term
12 January 2009 to 20 March 2009

Summer term
20 April 2009 to 26 June 2009