Learning and Knowledge in Social Networks
Robert Carrington

Abstract:
For most purposes, the information an agent can readily access is just
as important as the agent’s knowledge. This thesis explores several
approaches to reasoning about the information agents in a network can
access. The first section introduces a modality for information from
immediate connections, and axiomatizes the resulting epistemic logic
(EAL). I also introduce and axiomatize iterated version of the logic
(IAL), which considers information along multiple edges of access. I
prove both of these logics complete for epistemic models equipped with
an edge relation. I define and axiomatize two additional logics for
access without completeness – a version with restricted access (RAL)
and one extending the existing framework of epistemic friendship logic
with iterated access modalities (IFL).