vacancies for research assistants

[marina d'Engelbronner]

ROYAL HOLLOWAY

University of London

THREE RESEARCH ASSISTANT IN KERNEL BASED METHODS
Department of Computer Science

Royal Holloway, University of London invites applications for three
research assistant positions in Computer Science. The first post is
funded for two years duration, while the other two are funded for three
years. The starting date for the first project is flexible, but the
other vacancies are from April 2002.

The first vacancy is for an EPSRC funded project entitled, 'Delimiting
Kernel-based Learning Methods'. It spans work on the learning-theoretic
analysis of kernel-based methods, kernel-theoretic work, and algorithmic
design. A background working in the theoretical analysis of learning
methods is very desirable for this position.

Salary is up to =A324,656 inclusive of London Allowance Ref: KB/2053.

The 2 remaining vacancies are for a project that involves developing
kernel-based methods for the analysis of images, Learning for Adaptable
Visual Assistants. The project is financed by the EU and also involves
partners in Cambridge (Xerox), France (INRIA, Grenoble and University of
Rennes), Austria (Technical University of Graz), Switzerland (IDIAP) and
Sweden (University of Lund).  We are seeking researchers with experience
in applications of machine learning with a preference for knowledge of
image processing, and a strong programming background. Experience with
kernel methods is desirable but not required.  Salary is in the range
=A320,865 to =A328,625 per annum inclusive of London Allowance Ref:
KB/2052.

Please contact John Shawe-Taylor by email at jst at cs.rhul.ac.uk for more
information.

Information on the Department may be found at www.cs.rhbnc.ac.uk/. Further details and an application form can be obtained from The
Personnel Office, Royal Holloway, University of London, Egham, Surrey
TW20 0EX; fax: 01784 473527; tel: 01784 414241; email
Sue.Clarke at rhul.ac.uk Please quote the appropriate reference

The closing date for receipt of applications is 17th April 2002.

We positively welcome applications from all sections of the community



ROYAL HOLLOWAY

University of London

Department of Computer Science

Research Assistant - Kernal Based MethodsInformation for Candidates

=20
Royal Holloway, University of London invites applications for three
research assistant positions in Computer Science. The first post is
funded for two years duration, while the other two are funded for three
years. The starting date for the first project is flexible, but the
other vacancies are from April 2002.

The College
Royal Holloway is dedicated to achieving the highest international
standards in teaching and research in the Sciences, Social Sciences,
Humanities and Creative Arts.  The College exists to promote education
and scholarship for the public benefit.  By constantly reinterpreting
the ideals and principles of a university through undergraduate
teaching, advanced teaching and research we carry forward the vision of
our founders in the changing context of the society we serve.

Royal Holloway is one of the eight large Colleges of the University of
London. It has over 5,400 students in the Faculties of Arts, History and
Social Sciences and Science. There are 20 academic departments and
almost 1,200 staff, including 370 academic teaching staff.  As well as
strong links within the University of London, the College has forged
many successful national and international collaborations. There is a
strong commitment to excellence in both teaching and research and in the
1996 Research assessment Exercise, 8 departments were rated 5* or 5 and
none below 3.

The College occupies a large attractive campus at Egham, Surrey,
situated in the green belt near Runnymede and Windsor Great Park, with
good communications to and from London.  Egham is 35 minutes by train
from Waterloo, and the College is one mile from the M25 and 15 minutes'
drive from Heathrow Airport.

For further information about the College see website
http://www.rhbnc.ac.uk.

The Department

The Department of Computer Science and Computer Learning Research Centre
occupy the modern purpose-designed McCrea Building, which accommodates
staff offices, computing laboratories and terminal rooms, providing the
latest facilities for research and undergraduate teaching.  The
Department has 15 established academic posts in Computer Science,
expanding to 18 this year, and also employs 10 Research Assistants, 3
Technical support staff and 4 Administrative/secretarial staff. The Department was given a rating of 5 in the last (2001) Research
Assessment Exercise.

The Department's teaching load at present is around 170 Full Time
Equivalent (FTE) undergraduate students taking part in both Single and
Joint Honours Degree programmes, and around 35 FTEs, approximately half
taught and half research. The main taught postgraduate programmes are
the MSc in Computer Science by Research, which offers intensive training
in any one of the Department's research specialties, and the MSc in
Business Information Systems, taught jointly with the School of
Management.The Department is well equipped for teaching and research. It provides
networked computer facilities accessed from desktop X-terminals.  The
servers include powerful DEC Alpha computers running Digital Unix, a Sun
Sparc running Solaris, and an NT server running various office
applications such as Microsoft Office.  There is fast Internet access
(multi megabit per second), and a wide range of specialist research
software as well as laser printers and photocopiers.  All staff and
research postgraduates who do not have equipment provided under personal
research grants are provided with an X terminal.  This gives them access
to both Unix services and an NT server.

The Department holds weekly research seminars and invited lectures
throughout the academic year as well as a Distinguished lecture series -
recent speakers include Professors Donald Davies, Roger Penrose, Frank
Sumner, Maurice Wilkes and Tony Hoare.  The department celebrated its
30th anniversary in 1998.

The department's website is http:\\www.dcs.rhbnc.ac.uk

COMPUTER SCIENCE STAFF - RESEARCH INTERESTS

A. Gammerman, BSc, PhD St Petersburg (Head of Department)

Algorithmic randomness, Kolmogorov complexity, induction and
transduction, applications.

A. Chervonenkis, BSc, PhD Moscow

Mathematical statistics, pattern recognition, learning theory, ore
deposit modelling.

D.A. Cohen, BA, DPhil Oxon

Mathematics within Computer Science, theory of neural networks,
constraint satisfaction problems.

A.R. Davies, BSc, MSc Lond (Deputy Head of Department)

Computer modelling of large scale scientific problems, laser optics,
wave guides.

A. Fukshansky, Dipl Math, Dr rer nat Freiburg

Mathematics in computer science, bioinformatics/biomathematics.

Z.G. Gutin, BSc Gomel, PhD Tel Aviv

Graph theory and algorithms, combinatorial optimisation, linear and
integer programming, bioinformatics.

J.M. Hancock, BSc London, PhD Edinburgh

Bioinformatics, molecular evolution, genome analysis, repetitive
sequences.

E.I. Hogger, BSc, MSc Lond, BA Open, ARCS, DIC

Machine learning and its relationship to non-monotonic logic,
intelligent knowledge-based systems, expert systems.

A.I.C. Johnstone, BSc, PhD Lond, CEng, MBCS, MIEE

Multiprocessor systems for real-time machine vision, language design for
multiprocessor and array processor systems, VLSI implementation of image
processing algorithms, advanced processors.

C. Saunders, BSc, PhD Lond

Machine learning algorithms, kernel methods, transductive inference,
fault diagnosis, text analysis.S.A. Schneider, BA, DPhil Oxon

Process algebra, concurrency theory, real-time systems, formal methods,
computer security.

E.A. Scott, BSc, DPhil Oxon

Theoretical computer science, compiler theory, language analysis and
design, termination theory, automated theorem proving, machine learning.

J.S. Shawe-Taylor, PhD, MSc Lond, DIC, CMath, FIMA

Computational learning theory, the mathematical analysis of kernel-based
learning methods.

H.E. Treharne, BSc, MSc, PhD Lond

Safety critical software, formal methods, software metrics.

V.N. Vapnik,  BSc, PhD, DSc Moscow

Pattern recognition, statistical analysis, support vector machine.

V. Vovk, BSc, PhD Moscow

Limits of machine learning: predictive complexity, randomness and
information; inductive and transductive inference.

C. Watkins, MA, PhD Cantab

Reinforcement learning, computational learning theory, mathematical
finance.

RESEARCH

COMPUTATIONAL LEARNING RESEARCH CENTRE

The Centre was established in January 1998 to provide a focus for
fundamental research and commercial and industrial applications in the
fields of computer learning, including inductive/transductive inference
and universal prediction.  The current research topics are Universal
Prediction, Support Vector method, Probabilistic Reasoning, the theory
of Kolmogorov and predictive complexity, on-line prediction with expert
advice, transductive inference and computational finance.

Members of the Centre are Alex Gammerman (Director), Alexey
Chervonenkis, Craig Saunders, Vladimir Vapnik, Volodya Vovk and Chris
Watkins, with their Research Assistants and PhD students, and Visiting
Professors are Jorma Rissanen, Chris Wallace, Glenn Shafer, Leonid
Levin, Ray Solomonoff and Vladimir V'yugin.

THEORETICAL COMPUTING GROUP

The group is concerned with the modelling of computing systems and
application areas in order to derive principled and practically
effective solution strategies. Such modelling ensures that the
development of solutions is guided by a scientific and well-founded
analysis, which can provide practical guidance into the scaling and
extent of their applicability.  The research strands include:

NEURAL AND COMPUTATIONAL LEARNING

Members of this area work closely with the CLRC and their research is
centred around analysis of neural networks including work on a novel
digital neural chip, large margin algorithms and analysis, and relations
between the Bayesian and probably approximately correct (pac) models.
The research group has collaborated in ESPRIT funded research in the
mobile telecoms area and currently co-ordinates an ESPRIT funded Working
Group Neurocolt2. More recently they have received funding for an EU
project 'Kernel Methods for Images and Text' (KerMIT) involving Reuters,
Xerox and three other university sites.CONSTRAINTS

Constraint satisfaction problems have been studied in the Department
since 1989, supported by the EPSRC, the DTI, the Royal Society, the
British Council, and the Nuffield Foundation. The group is currently
collaborating with the government Radiocommunications Agency, and
Vodafone Ltd, who support a PhD student.  Theoretical research is
currently focused on developing the mathematical methods that are needed
to classify different types of constraints.  A new approach to
computational complexity theory, which makes use of tools from algebra
and logic, is being investigated.  On the applications side, radio
frequency planning particularly in regard to the growth of mobile
telecommunications, and large scale scheduling problems, such as the
planning of manufacturing processes and the design of airline
timetables, are of particular interest.

FORMAL METHODS

The main interests of the Formal Methods group concern the theory and
application of formal methods to security- and safety-critical systems,
with particular focus in the areas of concurrency and combining methods.
 The group has an international reputation in this field and is
substantially supported by external funding.  The group maintains active
links with industry, including Motorola, the Defence and Evaluation
Research Agency, and SRI.

LANGUAGES AND ARCHITECTURES

Research in the area of language translation and compiler theory aims to
provide sound theory linked to approachable toolsets for work in
software/hardware co-design, language implementation and reverse
engineering.  We have made contributions in the areas of general
parsing; semi-automatic construction of intermediate forms and control
flow graphs; control flow analysis for environments with poorly defined
notions of procedure call; reverse compilation from assembler source to
high level languages; mixed implementation of conventional processors,
Digital Signal processors and Field Programmable Gate Arrays; and the
development of toolsets for non-specialist users.

BIOINFORMATICS

The volume and variety of data coming out of current research in
molecular biology contain the answers to many scientific questions and
the keys to many medical advances.  To answer the new types of question
that are being asked, new computational techniques are needed, and
machine learning methods are now becoming standard in computational
biology.  Techniques developed in the Department are now being applied
to the analysis of biological sequences.  In particular, support vector
machine kernels suitable for the comparison and classification of 
protein and DNA sequences are being developed with a view to providing
predictions of the structure and function of proteins from their
sequences.

The Post
Delimiting Kernal-based learning
Assisting Professor John Shawe-Taylor with the above research project,
the aim of which is to analyse the limitations of kernal-based learning
methods.

Learning for Adaptable Visual Assistants=20
Assisting Professor John Shawe-Taylor with the above research project,
the aim of which is developing kernal based methods for the analysis if
images in collaboration with the other partners on the project.

General
Any other duties or responsibilities as the department may reasonably
require.

Opportunity to assist with undergraduate lectures and tutorials,
assistance with postgraduate supervision.

The PersonCandidates for all three positions will preferably have a doctorate in
Computer Science or Mathematics. They will be expected to be able to
demonstrate significant research accomplishment in a relevant field,
including published work in high quality journals.

Experience with kernal methods is desirable.

Enquiries
Please contact  John Shawe-Taylor by e-mail at jst at cs.rhul.ac.uk for
more information.

The Appointment
The first vacancy is for an EPSRC funded project entitled, 'Delimiting
Kernel-based Learning Methods'. It spans work on the learning-theoretic
analysis of kernel-based methods, kernel-theoretic work, and algorithmic
design. A background working in the theoretical analysis of learning
methods is very desirable for this position.

Salary is up to =A324,656 inclusive of London Allowance Ref: KB/2053.

The 2 remaining vacancies are for a project that involves developing
kernel-based methods for the analysis of images, Learning for Adaptable
Visual Assistants. The project is financed by the EU and also involves
partners in Cambridge (Xerox), France (INRIA, Grenoble and University of
Rennes), Austria (Technical University of Graz), Switzerland (IDIAP) and
Sweden (University of Lund).  We are seeking researchers with experience
in applications of machine learning with a preference for knowledge of
image processing, and a strong programming background. Experience with
kernel methods is desirable but not required.  Salary is in the range
=A320,865 to =A328,625 per annum inclusive of London Allowance Ref:
KB/2052.

The appointments will be made on 12 months probation. There is also an
annual appraisal scheme.

Payment for salaries will be monthly, in arrears, by credit transfer
into your bank account.  Payment will usually be made on the 27th of the
month or before if the 27th falls on a weekend or Public or College
holiday.  The normal date for the review of salaries is 1 April.  Where
a salary increment is due it will be payable on 1st August in each
calendar year.  Where a member of staff is appointed between 1st
February and 31 July inclusive the first increment will be payable on 1
August in the following calendar year.  This post is superannuable under the USS (Universities Superannuation
Scheme). The College currently contributes 14% of salary and individuals
contribute 6.35%.

27 days annual leave are attached to the post. In addition you will
receive the 6 discretionary days which are granted and shared between
Easter and Christmas when the College is closed and public holidays.

ApplicationsApplications should be made using the standard application form
obtainable from the Personnel Department (address below), to which
should be appended:

(I)         a full curriculum vitae;

(ii)        the names and addresses of two or more academic referees;

(iii)       a statement of current research activities and areas of
interest;

(iv)       a list of publications.

These should be sent by 17th April 2002  to Personnel, Royal Holloway,
University of London, Egham, Surrey TW20 0EX, UK (Tel: 01784 414241;
Fax: 01784 473527; E-mail: s.watson at rhbnc.ac.uk).Extract from equal opportunities policy
The only consideration in recruitment of employees will be how the
genuine requirements of the post are met or likely to be met by the
individual under consideration.  These requirements, including
retirement at the appropriate age, being met, no regard will be taken
(except where legally required) of that person's race, sex, age, marital
status, number of children, physical disability or beliefs or lawful
preferences, privately held, on any matter including religion, politics
and sex.

=20
Information for Applicants with Special Needs
Royal Holloway encourages and welcomes applications from people with
disabilities and special needs.  However it is important to note that
the Campus is built on a hill side and that many buildings are of older
construction and are not fully accessible.  If you have a disability or
a special need and wish to discuss any practical aspects of your
application, please contact the Personnel Office in confidence on +44
(0) 1784 414058.

General
In an effort to provide a healthy and comfortable working environment,
smoking is prohibited in public areas and in shared occupancy rooms. 
Full details of the Smoking Policy are available from the College Safety
Officer.

Staff have full use of the dining facilities on campus; there is also a
College Shop which offers a wide range of goods.  A National Westminster
bank is on site together with a Waterstones bookshop. There is an
independently run Nursery (for which a charge is made) located on the
College campus for children over the age of two years and up to the age
of five years.

Designated areas of car parking are allocated to staff. Permits are
required to allow parking on campus. A bus service is in operation from
Egham Railway Station and may be used by staff in addition to students. 
The service operates in conjunction with trains arriving from Waterloo
and Reading.  Tickets for this bus service must be bought in advance and
are available from the College Shop, Athlone Hall and Kingswood Hall. 
This bus service operates during term-time only.

Internal promotion is encouraged by the college and all vacancies are
advertised on internal noticeboards and circulated to departments.