{"id":"arxiv/2610.01099","url":"https://arxiv.org/abs/2610.01099","tags":["quant-ph"],"score":0,"title":"Constant-Per-Layer-Depth MPS-Pretrained Ansatz for Noisy Distributed Quantum Processors","domain":"quant-ph","source":"arxiv","authors":["Seongpyo Hong","Woodo Lee","Yong-Su Kim","Seung-Sup B. Lee","Junghyun Lee"],"reviews":0,"subject":"quant-ph","summary":"Distributed quantum processors could scale variational algorithms beyond single devices, but circuit depth, communication overhead, and noise limit their performance. We compare ladder, mixed-canonical, and brick-wall realizations of matrix-product-state (MPS) pretraining with matched per-layer two-qubit-block resources. Despite comparable ideal variational quantum eigensolver performance and gradient scales, the shallower brick-wall architecture reduces circuit duration, idle-time decoherence, and zero-noise-extrapolation (ZNE) overhead, yielding superior noisy and ZNE-assisted performance.…","category":"research","comments":0,"disputed":false,"division":"physics","mentions":0,"discussed":0,"published":"2026-10-01","reactions":{}}