Newsgroups: comp.lang.prolog,comp.ai,comp.ai.edu
Path: cantaloupe.srv.cs.cmu.edu!nntp.club.cc.cmu.edu!bb3.andrew.cmu.edu!goldenapple.srv.cs.cmu.edu!das-news2.harvard.edu!cam-news-feed3.bbnplanet.com!cam-news-hub1.bbnplanet.com!cpk-news-hub1.bbnplanet.com!news.bbnplanet.com!rill.news.pipex.net!pipex!join.news.pipex.net!pipex!server1.netnews.ja.net!warwick!bris.ac.uk!usenet
From: Christophe Giraud-Carrier <cgc@cs.bris.ac.uk>
Subject: MSc in ML at Bristol
X-Nntp-Posting-Host: uranus.cs.bris.ac.uk
Content-Type: text/plain; charset=iso-8859-1; name="Ad-txt.txt"
Message-ID: <33D5F125.AA6@cs.bris.ac.uk>
Content-Disposition: inline; filename="Ad-txt.txt"
Sender: usenet@fsa.bris.ac.uk (Usenet)
Content-Transfer-Encoding: 8bit
Organization: Department of Computer Science
Mime-Version: 1.0
Date: Wed, 23 Jul 1997 11:55:17 GMT
X-Mailer: Mozilla 3.01Gold (WinNT; I)
Lines: 198
Xref: glinda.oz.cs.cmu.edu comp.lang.prolog:17736 comp.ai:47611 comp.ai.edu:4233

**********************************
*                                *
*   MSc in Advanced Computing    *
*                                *
* University of Bristol, England *
*                                *
**********************************

**********************************
********                  ********
****     MACHINE LEARNING     ****
********                  ********
**********************************


Introduction:
=============

Machine learning is concerned with the design of algorithms that, 
rather than encoding explicit instructions or programs for the 
solution of specific problems, encode inductive mechanisms whereby 
solutions to broad classes of problems may be derived from examples. 
Whether the approach is inspired by biology (e.g., artificial neural 
networks) or by cognitive psychology (e.g., traditional AI), machine 
learning aims at building computer systems capable of adapting to 
new tasks and learning from their experience.

Machine learning has found increasing applicability in fields as 
varied as banking, medicine, marketing, condition monitoring, vision, 
programming and robotics. Skilled professionals and researchers, who 
can apply machine learning technology to current business problems 
and push the limits of what computers can effectively do further, 
are in demand.


Aim:
====

This new course is aimed at giving students a solid grounding in this 
exciting technology and at equipping them with the skills necessary to 
construct and apply ML tools to the solution of complex business 
problems. The course also provides the necessary foundation for PhD 
work in machine learning.


Objectives:
===========

At the end of this course, students will be able to:

- Determine and justify, given a problem, the suitability of machine 
  learning techniques to the solution of that problem
- Select, given a specific application, the most appropriate machine 
  learning technique to use
- Design and implement machine learning solutions to suitable problems
- Discuss the limitations of current approaches to machine learning
- Demonstrate creativity by suggesting ways to improve existing 
  techniques or developing new techniques and algorithms
- Communicate ideas and concepts clearly both orally and in writing


Entry Requirements:
===================

Applicants must have a good honours degree, or a similar qualification, 
in computer science or a closely related field. Knowledge of the 
following topics, though not essential, is strongly recommended: 
probability theory, logic and logic programming, discrete mathematics, 
and at least one mainstream procedural programming language (e.g., C/C++).

Students must also provide evidence of competence in English (e.g., 
IELTS if overseas).


Structure:
==========

This is a full-time course, starting in October and of twelve months 
duration. The course consists of lectures extending over two semesters, 
and colloquia and seminars extending over the twelve months of the course. 
During the final five months of the year, students undertake an original 
investigation in some aspect of machine learning, leading to a dissertation.

The workload is distributed as follows:

	Lectures:			160 hours (2 semesters)
	Individual Research/Work:	12+ hours per week (2 semesters)
	Project/Dissertation:		5 months (May-September)	

Students are expected to attend lectures and seminars and to occasionally 
give seminars, such as review of a paper or presentation of a topic related 
to their project. Taught units, often supplemented by seminars from invited 
speakers, cover the following topics:

- Introduction to artificial intelligence
- Propositional inductive learning
- Neural networks
- Computational intelligence (e.g., fuzzy systems)
- First-order inductive learning (e.g., ILP)
- Soft computing (e.g., genetic algorithms, artificial life)
- Advanced topics in machine learning (e.g., hybrid systems)
- Computational learning theory

The final project is a substantial piece of work, which may be commercially 
or industrially related. Joint projects with local industries are often 
available and encouraged. Topics for projects are offered by both staff 
members and industrial partners. Hence, topics tend to reflect the current 
research interests of the Department and the needs of industry. Students 
may also suggest their own topic to staff members for supervision. Machine 
learning-related projects carried out by final year and MSc students have 
included:

- Design and implementation of a mail agent for Pine
- Application of neural networks to speech processing
- Genetic evolution of Othello-playing neural networks
- Web-navigating and searching agents
- Evolution of neural networks using genetic algorithms and lamarckism
- Complex adaptive systems and artificial life
- Evolutionary optimisation of RBF networks
- Improvement of rule selection in expert systems using neural networks
- Incremental learning, recurrent neural networks and NLP


Assessment:
===========

Assessment of the taught part of the course is by practical work and 
examination. Practical work consists of report writing, oral presentations, 
exercises and simulations. Some units may be assessed entirely on practical 
work. Examinable course units are examined in May/June. Candidates must 
achieve an average mark (practical work and examination) of 40% to proceed 
to the project.

Project work carried out during the final part of the course is reported in 
a written dissertation and demonstrated to the supervisor. The dissertation 
is assessed by members of staff from the Department, and the resulting 
assessment is moderated by an external examiner. The pass mark for the 
dissertation is 40%.

Successful completion of the course leads to the award of the MSc degree. A 
commendation may be awarded for an excellent performance in both the taught 
part of the course and the project work. A candidate who fails to meet the 
required standard of performance for the award of the MSc may be awarded a 
Diploma.


Financial Commitment:
=====================

Fees for the 1997/1998 sessions are:

- Home/EC students	2540 per annum
- Overseas students	8798 per annum

It is estimated that about 6000 per annum is required for subsistence in 
Bristol for a single person, 8000 for a couple, with an additional 700-1100 
for each child.

Overseas applicants may be able to obtain a scholarship from the British 
Council. Please approach the Councillor in your home country directly for 
details.

For UK applicants, career development loans are available to students from 
commercial banks. These loans are normally interest-free for the first 13 
months with repayment of the loan starting in the fourteenth month. A number 
of students finance their studies in this way.

The Department continually seeks methods of supporting students with 
scholarships. Therefore, please enquire at the time you apply as the 
availability of scholarships may have increased since your initial enquiry.


Career Prospects:
=================

This course offers suitable training and experience for graduates to 
undertake employment in research establishments, industry and academia. 
Graduates with the skills developed on this course are in demand from 
traditional employers, as well as more modern businesses. The existence of 
large legacy databases and the accelerating production and broad dissemination 
of data mean that there is an increasing need for automatic mechanisms to 
filter, mine and summarise information.  
 

Important Information:
======================

The course starts in October. The number of places on the course is limited. 
Application forms are available from Mrs Ann Prowse, Department of Computer 
Science, University of Bristol, BS8 1UB, England. Tel: +44-117-954-5130. 
Email: ann@cs.bris.ac.uk.

For more information on the course and its contents, please contact Christophe 
Giraud-Carrier by email at: cgc@cs.bris.ac.uk.

For information on current research interests in machine learning at Bristol, 
feel free to visit our web site at http://www.cs.bris.ac.uk/Research/MachineLearning.

