Perfect cryptography, S5 knowledge, and algorithmic knowledge

  • Authors:
  • Sabina Petride;Riccardo Pucella

  • Affiliations:
  • Oracle Corporation;Northeastern University

  • Venue:
  • TARK '07 Proceedings of the 11th conference on Theoretical aspects of rationality and knowledge
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

We propose a principled approach to model secrecy in multiagent systems, by defining a set of possible observations and providing agents with algorithms used to distinguish the possible states of the system. Our approach fits naturally within a knowledge-based account of secrecy. By adjusting both the kind of observations and the capabilities of the agents, we can capture in a natural way different forms of secrecy in the presence of perfect cryptography. In particular, we show how to model extraction secrecy. Our formalization suggests a unified definition of secrecy for cryptographic protocols and for systems that seek to prevent inadmissible flows of information.