A Scalable Quantum Neural Network for Approximate Unitary Synthesis

In this work, scalable quantum neural networks are introduced to approximate unitary evolutions through the Standard Recursive Block Basis (SRBB) and, subsequently, redesigned with a reduced number of CNOTs. This algebraic approach to the problem of unitary synthesis exploits Lie algebras and their topological features to obtain a continuous parameterization of unitary operators. First, we implemented the recursive construction of the SRBB to provide a starting tool for quickly verifying the mathematical properties needed to guarantee the original scalability scheme, already known to the literature only from a theoretical point of view. Unexpectedly, 2-qubit systems emerge as a special case outside this scheme. Furthermore, we present a method to reduce the number of CNOT gates, thus deriving a new implementable scaling scheme, which requires only one single layer and whose performance has been tested with a variety of different unitary matrices via the PennyLane library.