Computing Science

University of Stirling

Spring 1999 Seminars

The Department of Computing Science and Mathematics presents the following seminars. Unless otherwise stated, seminars will take place in Room 4B94 of the Cottrell Building, University of Stirling from 15.00 to 16.00.

If you would like to give a seminar to the department in future or if you need more information, please contact the seminar organiser Ken Turner (Phone 01786-467-420, Email kjt@cs.stir.ac.uk).


17th February 1999

Development of a Validation System based on Formal Description Techniques: Application to an Ada Run-Time System

Dr. Tomas Robles, Polytechnic University of Madrid

The validation of the ATAC (Ada Tasking Coprocessor) chip will be covered. The general approach will be described, along with the environment and tools used for validation. The ATAC requirements and their formal specification in LOTOS will be outlined. Results will be presented from conformance testing of the ATAC system.

5th March 1999

Proofs by Pictures

Dr. Sharon Curtis, University of Stirling

The title refers to a graphical calculus, which is a mathematical tool allowing formulae to be represented and reasoned about using a visual representation. A formula may be represented using a graph, and there are laws for transformation of graphs, with corresponding effects on the formulae represented. Using graphs makes formulae easier to understand, and often makes the next step in a proof more obvious.

The use of this calculus for various computational applications will be explained, including the development of algorithms and systems of sequential processes.

This talk should be of interest to anyone in Computing Science or Mathematics, from final-year undergraduate level upwards.

19th March 1999

SDL in Rigorous Object-Oriented Analysis

Dr. Robert Clark, University of Stirling

We will demonstrate how SDL can be applied during object-oriented analysis to produce a formal object-oriented requirements specification. Building a formal specification from informal requirements is difficult. To simplify this, we first use a scenario-based approach to build a formal user-centred model that specifies behaviour from the viewpoint of the users. The user-centred model is used to help create the formal object-oriented specification. Both the user-centred model and the object-oriented specification are represented in SDL. We validate the user-centred model with respect to the requirements. Validation of the user-centred model can then be achieved by demonstrating that it provides the behaviour expected by the user-centred model.

26th March 1999

Theorem Proving at Full Throttle

Helen Lowe, Glasgow Caledonian University

The role of proof in formal systems development is often neglected in undergraduate courses on formal methods. The point of formal methods is to find mistakes, clear up misunderstandings, and resolve ambiguities.

The idea behind this work is the "Soccer referee's book" by Yves Ledru. The general idea is interesting but needs extension. Speedway, on the other hand, is full of ambiguities, contradictions, missing cases and sheer incomprehensibility. Moreover, the rules are constantly changing in the quest to achieve both justice and excitement, giving even more scope for faulty specification and eleventh-hour `patches'.

In this talk I will give examples of possible rule violations, ambiguities, and incompleteness. I will show how a formalisation might be of help (but to not the average speedway referee, unless we automate the process ... but that is another story.)

16th April 1999

Neuromorphic Engineering

Dr. Catherine J. Breslin , University of Stirling

Neuromorphic engineering uses the technology of silicon integrated circuits to construct models for developing and testing theories of the brain and to build replications of brain function.

This talk will present some of the significant advances (and setbacks) in the history of brain modelling and brain replication, and how neuromorphic engineering fits into the picture.

The original motivation for using silicon integrated circuits was the concept of a physical equivalence between field-effect transistors and the channels that allow ions to flow through the membranes of neurons. However, this is mostly neglected in the field today, which is strongly biased towards brain replication, rather than brain modelling.

The talk will be peppered with examples, including: silicon neurons, quadruped locomotion and robotics, whole behaving animals, evolving silicon and interfaces between transistors and brain cells.


Up one level to Computing Science Seminars

Web Ken Turner Home   Email    Search Search Web Pages


URL: http://www.cs.stir.ac.uk/~kjt/seminars/spr99.html