Causal structures and the classification of higher order quantum computations

Abstract

Quantum operations are the most widely used tool in the theory of quantum information processing, representing elementary transformations of quantum states that are composed to form complex quantum circuits. The class of quantum transformations can be extended by including transformations on quantum operations, and transformations thereof, and so on up to the construction of a potentially infinite hierarchy of transformations. In the last decade, a sub-hierarchy, known as quantum combs, was exhaustively studied, and characterised as the most general class of transformations that can be achieved by quantum circuits with open slots hosting variable input elements, to form a complete output quantum circuit. The theory of quantum combs proved to be successful for the optimisation of information processing tasks otherwise untreatable. In more recent years the study of maps from combs to combs has increased, thanks to interesting examples showing how this next order of maps requires entanglement of the causal order of operations with the state of a control quantum system, or, even more radically, superpositions of alternate causal orderings. Some of these non-circuital transformations are known to be achievable and have even been achieved experimentally, and were proved to provide some computational advantage in various information-processing tasks with respect to quantum combs. Here we provide a formal language to form all possible types of transformations, and use it to prove general structure theorems for transformations in the hierarchy. We then provide a mathematical characterisation of the set of maps from combs to combs, hinting at a route for the complete characterisation of maps in the hierarchy. The classification is strictly related to the way in which the maps manipulate the causal structure of input circuits.