Factorizing time evolution into elementary steps

TL;DR

This paper introduces a novel method to decompose quantum system time evolution into a sequence of elementary, time-ordered operations using optimization techniques, enhancing control and implementation flexibility.

Contribution

It presents a new algorithm that optimizes the durations, driving functions, and elementary operations for quantum time evolution factorization, building on Lie algebra methods.

Findings

Produces optimal sequences of unitary operations

Allows relaxation of key assumptions for practical use

Provides a comparison with existing factorization approaches

Abstract

We propose an approach to factorize the time-evolution operator of a quantum system through a (finite) sequence of elementary operations that are time-ordered. Our proposal borrows from previous approaches based on Lie algebra techniques and other factorization procedures, and requires a set of optimization operations that provide the final result. Concretely, the algorithm produces at each step three optimal quantities, namely the optimal duration of the desired unitary operation, the optimal functional dependence of the driving function on the optimal time, and the optimal elementary Hermitian operation that induces the additional unitary operation to be implemented. The resulting sequence of unitary operations that is obtained this way is sequential with time. We compare our proposal with existing approaches, and highlight which key assumptions can be relaxed for practical implementations.