 |
|||||||||||
| Home | Admissions | Students | Careers | Research | Business | People | Help | ||||
 |
|
|
|
|
|||||||
|
|
|
 |
||
 |
||
|
| Getting Here | | Contact | |
||
 |
||
|
|
||
|
News  Events Listing Events - Additional Information Vacancies Staff Intranet |
||
|
|
|
| HOME > Event Information | CSML Seminar: Biologically-Inspired Massively-Parallel Architectures - computing beyond a million processors
Speaker: Steve Furber, School of Computer Science, University of Manchester, UK AbstractThe SpiNNaker project aims to develop parallel computer systems with more than a million embedded processors. The goal of the project is to support large-scale simulations of systems of spiking neurons in biological real time, an application that is highly parallel but also places very high loads on the communication infrastructure due to the very high connectivity of biological neurons. The scale of the machine requires fault-tolerance and power-efficiency to influence the design throughout, and the development has resulted in innovation at every level of design, including a self- timed inter-chip communication system that is resistant to glitch-induced deadlock and \'emergency\' hardware packet re-routing around failed inter-chip links, through to run-time support for functional migration and real-time fault mitigation.CS Internal Seminar Series: Tim Weyrich and Jens Groth
UCL Contact: Melanie Johnson
(Visitors from outside UCL please email in advance). AbstractTim Weyrich: \"Authenticating Blank Paper Using Commodity Scanners\"Jens Groth: \"Efficient Zero-Knowledge Proofs\" CSML Seminar: The error in ABC
Speaker: Richard Wilkinson, University of Nottingham AbstractApproximate Bayesian computation (ABC) methods are a class of approximate algorithms for parameter estimation using only simulations from a computer model, and are increasingly being used in many different application areas. The quality of the approximation can be considered in terms of performing inference in the presence of added noise. We can control the distribution of this noise and use it to represent model error and/or measurement error. In this talk I will show how we can control this error and suggest how it should be used to move from making approximate statements about the model, to hopefully making consistent statements about the system.UCL-CS, UCL SECReT Seminar: Economics and the Future of IT Security
Speaker: Bruce Schneier
Please register at http://www.cscs.ucl.ac.uk/events-1/bruce-schneier-event/bruce-schneier See also http://www.cscs.ucl.ac.uk/events-1/bruce-schneier-event/bruce-schneier AbstractSurveying current trends in information security, it\'s clear that a myriad of forces is at work. But fundamentally, security is all about economics: both attacker and defender try to maximize the return on their investments. Understanding economics is critical to understanding IT security; it explains why security fails so often, and can offer new solutions for security success. For example, often the people who could protect a system are not the ones who suffer the costs of failure. Changing those economic incentives will do more to improve security than additional technology.Bruce SchneierBruce Schneier is an internationally renowned security technologist, described by The Economist as a security ‘guru’. He is the author of eight books -- including the best sellers Beyond Fear: Thinking Sensibly about Security in an Uncertain World, Secrets and Lies, and Applied Cryptography. His influential newsletter Crypto-Gram, and his blog Schneier on Security, are read by more than 250,000 people. He is a frequent guest on television and radio, and is regularly quoted in the press on issues surrounding security and privacy. Schneier is the Chief Security Technology Officer of BT. |
|
|
|
|
|||
| This page last modified: 4 December, 2007 by Graham Knight | |||
Computer Science Department - University College London
- Gower Street - London - WC1E 6BT - 
+44 (0)20 7679 7214 - Copyright © 1999-2007
UCL
