Structure-Fair Quantum Circuit Complexity: An Information-Theoretic Lower Bound

HongZheng Liu, YiNuo Tian, Zhiyue Wu

公開日: 2025/9/20

Abstract

Fairly quantifying the complexity of quantum states that possess intrinsic structures (such as symmetries or encodings) is a fundamental challenge for benchmarking quantum technologies. We introduce the "Reference-Contingent Complexity" (RCC). Its core idea is to utilize the quantum relative entropy to measure the extent to which a state deviates from its "structured vacuum" (the maximum entropy state within its constrained subspace), thereby only pricing the generation of non-trivial information. We establish a central theorem, establishing that the RCC is a rigorous information-theoretic lower bound on universal quantum circuit complexity, a bound that features a linear leading term, a universal logarithmic correction, and a precise physical correction for the non-uniformity of the spectral distribution. Furthermore, we establish an operational protocol based on quantum hypothesis testing, which connects this theoretical lower bound to experimentally accessible observables. This work provides a structurally-fair yardstick for quantum technologies and offers a new perspective for exploring the intrinsic connection between computational cost and the generation of non-trivial information.