Research Member of Common Room
Department of Computer Science
My research focusses on the relationship between trust and security. My methodology is grounded in formal methods, meaning I want to prove properties of systems or approaches.
Recently, I have been studying attacks on trust and reputation. Malicious participants (attackers) attempt to subvert the mechanisms of a trust or reputation system to gain an unfair advantage. Can we harden a system to an extent where we can prove that the damage introduced by attackers is limited?
We use techniques from information theory to reason about ratings that potentially are sent by attackers. Moreover, we investigate methods of using information in a way that is provably robust against attacks. We also use techniques from probabilistic modelling (Markov processes) to verify bounds on the probability of an attacker breaking the system.