The entropy of observables on quantum logic (Q1074578)

Let L be a quantum logic, i.e., L is an orthocomplemented, orthomodular \(\sigma\)-lattice. Let A be an observable (i.e., \(\sigma\)-homomorphism from the Borel subsets \(B({\mathbb{R}}_ 1)\) into L), and let \(\alpha\) be a state, i.e., a probability measure on L. The resolution of 1 with respect to an observable A is a set \(M\subset \{A(E):\) \(E\in B({\mathbb{R}}_ 1)\}\) with countably many elements \(\{a_ i\}\) such that \(a_ i\perp a_ j\), \(i\neq j\), \(\bigvee_{i}a_ i=1\). The set of all resolutions is denoted by \(m_ A\). Then the information of an experiment (resolution) \(M=\{a_ i\}\in m_ A\) in the state \(\alpha\) is given by \(H(\alpha,M)=- \sum_{i}\alpha (a_ i) \ln \alpha (a_ i)\). The entropy of an observable in a state \(\alpha\) is \(H(\alpha,A)=\sup_{M\in m_ A}H(\alpha,M).\) The author proves some of basic properties of the entropy. He shows that infinite entropy may be approximated by finite ones. The entropy of two experiments corresponding to compatible observables is defined, too, and it is shown that the entropy is the subadditive function. Finally, some remarks mainly on the logic of a Hilbert space are made.