Location: CLT 7.01 Date/Time: 3rd March 2015, 14:00 - 15:00
Speaker(s): Professor Kevin Jones (Plymouth University).
Brief Description:Many 'good practices' in computer security are based on assumptions and local evidence that do not generalize. There are few quantifiable methods of establishing or refuting the validity of these practices from a user perspective. We propose a formal model of security policies that allows us to evaluate the claimed benefits to the user of the system quantitatively. We illustrate the use of the model by looking at a security policy we all live with daily: The Password Policy, and show that some common practices are not well founded.
Location: CLT 7.01 Date/Time: 6th March 2015, 14:00 - 15:00
Speaker(s): Keith Clark.
Brief Description:This talk will briefly introduce the language features of TeleoR, a major extension of Nilsson's Teleo-Reactive procedures language. It is a rule based language in which the earliest rule in the sequence of rules of each called procedure, which has a guard that is inferable from the most recent set of percept facts, is fired. This becomes the call's fired rule. The action is either a set of robotic actions, executed in
parallel, or a call to a TeleoR procedure. A computation state has a stack of procedure calls with a record of which rule was fired, with what instantiation, for each call. At the top of the stack is a rule firing with primitive robot actions......
Location: Room M4.13, Merz court Date/Time: 16th March 2015, 14:00 - 15:30
Speaker(s): Andrey Mokhov.
In this talk I will give a brief report about my recent work at Microsoft Research Cambridge. Firstly, I will talk about the development of a new build system for Glasgow Haskell Compiler, which involved reasoning about graphs, concurrency and uncertainty. In the second part of the talk, I will discuss a preliminary study on compact models for Concurrent Kleene Algebra, a formalism introduced by Tony Hoare that unifies various laws of concurrent programming. After the talk, I would like to start a discussion about 'model compactness', looking into the following questions: i) is this a useful/meaningful notion? ii) when is compactness beneficial? iii) how can we achieve compactness?