Categorical quantum mechanics is the study of quantum foundations and quantum information using paradigms from mathematics and computer science, notably monoidal category theory. The primitive objects of study are physical processes, and the different ways that these can be composed. It was pioneered in 2004 by Abramsky and Coecke.
Mathematically, the basic setup is captured by a dagger symmetric monoidal category: composition of morphisms models sequential composition of processes, and the tensor product describes parallel composition of processes. The role of the dagger is to assign to each state a corresponding test. These can then be adorned with more structure to study various aspects, including:
A substantial portion of the mathematical backbone to this approach is drawn from Australian category theory, most notably from work by Kelly and Laplaza, Joyal and Street, Carboni and Walters, and Lack.
One of the most notable features of categorical quantum mechanics is that the compositional structure can be faithfully captured by a purely diagrammatic calculus.
These diagrammatic languages can be traced back to Penrose graphical notation, developed in the early 1970s. Diagrammatic reasoning has been used before in quantum information science in the quantum circuit model, however, in categorical quantum mechanics primitive gates like the CNOT-gate arise as composites of more basic algebras, resulting in a much more compact calculus.
One of the main successes of the categorical quantum mechanics research program is that from seemingly very weak abstract constraints on the compositional structure, it was possible to derive many quantum mechanical phenomena. In contrast to earlier axiomatic approaches which aimed to reconstruct Hilbert space quantum theory from reasonable assumptions, this attitude of not aiming for a complete axiomatization may lead to new interesting models that describe quantum phenomena, which could be of use when crafting future theories.