Dynamic Evidence Logics with Relational Evidence Andrés Occhipinti Liberman Abstract: Dynamic evidence logics are logics for reasoning about the evidence and evidence-based beliefs of agents in a dynamic environment. This thesis develops a family of dynamic evidence logics which we call relational evidence logics (REL). Relational evidence logics aim to contribute to the existing work on evidence logics in three main ways. First, while existing evidence logics model pieces of evidence as sets of possible states, REL models represent pieces of evidence as evidence relations. Evidence relations order states in terms of their relative plausibility, given a specific observation or instance of communication. Second, REL models include a representation of the relative reliability of the available pieces of evidence. This additional structure in the models is used to study reliability-sensitive forms of evidence aggregation. Third, various evidence aggregators are explored, to model alternative policies of the agent towards combining evidence.