Geometric Complexity of Quantum Channels via Unitary Dilations

Abstract

Nielsen's geometric approach to quantum circuit complexity provides a Riemannian framework for quantifying the cost of implementing unitary closed-system dynamics. For open evolutions, the reduced dynamics is described by time-dependent families of quantum channels $\{{\Lambda }_t{\}}_{t\ge 0}$ admitting many inequivalent Stinespring realizations, so any meaningful complexity notion must specify which microscopic resources are counted and which transformations are treated as gauge. We develop a dilation-based geometric complexity framework that distinguishes implementation-dependent complexity, attached to a specific realization, from intrinsic channel complexity obtained by minimizing over a physically admissible class of dilations. The proposed functional is subtractive: it removes purely environmental contributions from the geometric cost of the joint unitary realization. The subtraction is derived from concise structural postulates --- closed-system consistency, environment-only neutrality, and invariance under dilation gauge transformations --- and admits a natural variational and geometric interpretation. We also define noise complexity as a companion quantity measuring the geometric deviation of a noisy implementation from a prescribed ideal closed-system evolution. In the Hilbert-Schmidt setting, we establish coherence-based lower bounds for unitary and noise complexity, prove linear time scaling under time-homogeneous dilation models, and obtain dissipator-controlled bounds in the Markovian GKSL regime. We illustrate the framework on canonical single-qubit benchmark channels: dephasing, amplitude damping, and Pauli/depolarizing noise.