Lab Lunch Talks 2002
Archive: 1994 · 1995 · 1996 · 1997 · 1998 · 1999 · 2000 · 2001 · 2002 · 2003 · 2006 · 2007 · 2008 · 2009
November, 5th, 2002 Markus Frick < mfrick@inf.ed.ac.uk >
Primes in P I will shortly review the most important algorithms to deciding if a number is prime or composite. After mentioning the coNP, NP and the randomized polynomial time algorithms, I will briefly explain how this new determistic polynomial time algorithm works (no proofs).October 8th, 2002 Fedor Fomenko < F.Fomenko@sms.ed.ac.uk >
Report on CSL'02 September 24th, 2002 Massimo Felici < massimo.felici@ed.ac.uk >
On the Dependable Information Society: Events Report I will report about few events I have recently attended.
- ECCE-11 & SAFECOMP 2002, 8-13 September, Catania, Italy. I will quickly show the programme of the two conferences emphasizing the overall message.
- AMSD, DDSI, ACIP workshop, 18-20 September, Brussels, Belgium.
The main part of my talk will focus on one of the Keynote talks at SAFECOMP 2002:
"Towards a dependable Information Society: from DEPPY to FP6", Andrea Servida (European Commission)
Servida's talk is the intersection of the events listed above. I will go through Servida's slides to discuss about the new instruments within the EU FP6. The talk might be boring for who already know all the above things, but interesting for understanding the EU vision on research (funds). The talk will be really disappointing for who is looking for technical content...
September 3rd, 2002 John Longley < jrl@dcs.ed.ac.uk >
Some half-baked thoughts on the consciousness problem I will briefly survey some possible positions on the nature of the human mind and consciousness. Dennett and others have eloquently defended the view that in consciousness can in principle be explained in purely conventional scientific terms. I will present an outline an argument which appears (to me) to weigh against this position, and which also appears (to me) to be at least partly new. Friendly discussion of alternative points of view will be welcomed!
August 20th, 2002 Paul jackson < pbj@dcs.ed.ac.uk > and Ian Stark < Ian.Stark@ed.ac.uk >
FLoC '02 August 13th, 2002 Jane Hillston < jeh@dcs.ed.ac.uk >
PEPA for life July 30th, 2002 Kirill Morozov < kirill@brics.dk >
Unconditionally Secure Oblivious Transfer Protocols Based on Noisy Channels Oblivious Transfer (OT) is a cryptographic primitive which allows one party, traditionally called Alice, to transmit two secrets to another party, Bob, in such a way that Bob can choose to see one but only one of two secrets, while at the same time Alice does not know which secret is chosen by Bob. Despite that this primitive looks rather harmless it in fact allows us to solve many important problems in cryptography, e.g. it allows us to construct multiparty communication protocols as well as zero-knowledge proofs. In the talk, I will sketch the OT protocol by Crepeau '97, briefly argue why the protocol is secure, present possible generalisations and discuss some open problems.
July 16th, 2002 Linda Brodo < brodo@sci.univr.it >
UML and pi-calculus: another translation The talk will be about a first attempt to translate a UML-collaboration diagram into pi-calculus. The goal is to analyse some performance measures collecting information from the UML-model. An interesting aspect is that we do not use a stochastic process algebra, but the classic pi-calculus. Another point that will be shown is the relation between an UML message and a pi-calculus synchronisation. The work is still going on and comments and suggestions will be appreciated.
July 9th, 2002 Alan Bundy < bundy@dai.ed.ac.uk >
Informatics Research Methodologies When we launched the AI Research Methodologies (AIRM) module in 2000/01, it was on the understanding that this would be extended to include Cognitive Science and then Computer Science, eventually to become an Informatics Research Methodologies (IRM) module. Mike Ramscar carried out the first part of this programme in 2001/02 with his Experimental and Analytic Research Methodologies (EARM) module. I now wish to complete the programme with the inclusion of Computer Science.
At first sight, it looks as if many of the ideas in AIRM can be readily translated to the CS context. Checking this bold assertion is the point of my talk. I hope people will come along to confirm/deny this and to give me some new ideas.
July 2nd, 2002 Mark Jerrum < mrj@dcs.ed.ac.uk >
Spin models with two states June 25th, 2002 Monika-Jeannette Lekuse < mkl@dcs.ed.ac.uk >
(A Visit To South Bridge) May 28th, 2002 Colin Stirling < cps@dcs.ed.ac.uk >
Director's report for LFCS May 14th, 2002 Julian Bradfield < jcb@dcs.ed.ac.uk > and Perdita Stevens < pxs@dcs.ed.ac.uk >
Report on EPSRC Advanced Fellows' Seminar May 7th, 2002 Robert Atkey < roba@dcs.ed.ac.uk >
The TYPES 2002 Workshop in Nijmegen, 24th - 28th April April 30th, 2002 Ulrich Schöpp < us@dcs.ed.ac.uk >, Fedor Fomenko < F.Fomenko@sms.ed.ac.uk > and Thomas Ridge < T.J.Ridge@sms.ed.ac.uk >
Report on the Spring School in Semantics April 23th, 2002 Gonzalo Genova < ggenova@inf.uc3m.es >
Mapping UML Associations into Java Code UML stands for Unified Modeling Language, the "de facto" standard language for object oriented analysis and design of software systems. One of the key concepts of the language is "association", which is a type of connection between classes of objects. Associations are used to define structural and functional relationships between objects. Many CASE Tools in the market are able to generate code for different programming languages starting from UML models, but usually they are limited to skeletal code of classes, attributes and methods, without an adequate treatment of associations. We are working on a tool that will generate Java code to implement binary associations, having into account the features of multiplicity, navigability and visibility.
April 16th, 2002 Martin Lange < martin@dcs.ed.ac.uk >
ETAPS '02 April 2nd, 2002 Markus Frick < mfrick@dcs.ed.ac.uk >
Report on STACS 2002 March 26th, 2002 Bartek Klin < klin@brics.dk >
Charles Babbage: the grandfather of computing March 19th, 2002 Wenfei Fan < wenfei@research.bell-labs.com >
DTD-Directed Transformations for Data Exchange in XML For data exchange on the Web, e.g., in B2B-applications and the health-care industry, it is typical to transform relational data into an XML format that conforms to a predefined DTD (type). The presence of recursion in a DTD, among other things, makes it challenging to generate DTD-directed, efficient transformations. In this informal talk I shall present some half-baked ideas to develop a DTD-directed transformation system, based on a notion of attribute transformation grammars. The talk will pose more questions than answers.
March 12th, 2002 Martin Grohe <grohe@dcs.ed.ac.uk>
The Unusual Effectiveness of Logic in Computer Science March 5th, 2002 Marek Karpinski <marek@cs.uni-bonn.de>
On Some Certificate Problem for an Existence of a Perfect Matching We formulate and discuss an algorithm for a short cerification of a perfect matching, a problem arising in parallel and distributed computing. Some side-effects of this algorithm in development of sparse algebra algorithms, and design of efficient arithmetic circuits will be also discussed. In that context we formulate an intriguing open question.
February 26th, 2002 Dave Robertson < dr@dai.ed.ac.uk >
Open Problems in Agent Systems I shall present some personal thoughts on the way agent systems research has developed. In particular, I shall suggest some open problems where people with a talent for applications of formal methods could make a contribution.
February 19th, 2002 Jon Cook < jjc@dcs.ed.ac.uk >
Reverse Execution of Java Bytecode I have implemented a debugger for Java with a facility to run Java programs backwards at the non-native level. I will talk about how this is achieved for different bytecode instructions and how it is possible to support multi-threading, exceptions and garbage collection.
January 29th, 2002 Julian Bradfield < jcb@dcs.ed.ac.uk >
Dagstuhl seminar on "The Logic of Rational Agency" January 22th, 2002 Jennifer Tenzer < J.N.Tenzer@sms.ed.ac.uk >
A Formal Semantics of UML Class Diagrams based on Transformation Systems I'm going to talk about how a formal semantics for UML class diagrams can be defined in terms of transformation systems. The latter extend labelled transition systems and provide a common semantic domain for different diagram types included in the UML.
The background for this talk is my master thesis "A formal semantics of UML class diagrams based on transformation systems", which is available as technical report at
http://www.cs.tu-berlin.de/cs/ifb/Ahmed/RoteReihe/2001/TR01-9.ps
for those who want to have a closer look.
Archive: 1994 · 1995 · 1996 · 1997 · 1998 · 1999 · 2000 · 2001 · 2002 · 2003 · 2006 · 2007 · 2008 · 2009