Probabilistic Disjunctive Normal Forms in Temporal Logic and Automata Theory
This work addresses the challenge of effectively modeling and reasoning about uncertainty within temporal logic and automata theory. It proposes Probabilistic Disjunctive Normal Form (PDNF), which encodes the probabilities of a variable’s presence, absence, or negation through real-valued weights, and constructs a PDNF vector space to enable algebraic evidence fusion. Innovatively integrating probability distributions with venjunction structures from temporal logic, the approach establishes a Banach space framework that unifies logical semantics with functional-analytic properties. Leveraging exponential parameterization and tools from functional analysis, the study demonstrates that PDNF addition is equivalent to Bayesian evidence fusion and derives probabilistic bounds for identifying outcomes from random samples.