Curator's Take
AI Commentary
This article shows that a carefully engineered tensor‑network contraction can classically reproduce the output of IBM’s 70‑qubit doped Clifford sampling experiment in under an hour on a modest GPU cluster, cutting the required memory by two orders of magnitude compared with IBM’s own estimate. By proving that the contraction width ⌈d⁄2⌉ is optimal for these brickwork circuits, the authors demonstrate that even circuits with hundreds of T‑gates remain tractable for classical simulation when the entangling structure is one‑dimensional. The result sharpens the benchmark for quantum‑advantage claims and signals that future supremacy demonstrations will need deeper, more highly connected circuits to stay beyond the reach of state‑of‑the‑art classical algorithms.
— Mark Eatherly
Summary
We classically simulate the IBM doped Clifford random circuit sampling experiment, comprising $70$ qubits, $70$ entangling layers, and $468$ inserted $T$ gates. A deterministic temporal-boundary tensor network contraction approach is specifically designed to tackle such open-boundary one-dimensional brickwork circuits with operator-Schmidt-rank-$2$ entangling gates. For an $n$-qubit circuit of depth $d$, the resulting unsliced path evaluates an exact amplitude with contraction width $\lceil d/2\rceil$; Ratcatcher calculations certify that no smaller width is possible for the tested instances. Because one-qubit gates are absorbed without changing the network topology, the width and dense scheduled contraction cost are independent of their values and of the number and placement of $T$ gates. For the IBM instance, its largest intermediate tensor contains $2^{35}$ complex64 entries (256 times smaller than IBM's estimation), corresponding to a tensor payload of $256$ GiB, and is distributed across eight GPUs within a node. Using 32 nodes, with eight NVIDIA H100 GPUs per node, we completed all 2051 amplitude batches corresponding to IBM's published output bitstrings in 37.3 minutes. The resulting probabilities yield a log-XEB estimate of $0.35034$ with a 95\% interval of $[0.29763,0.40305]$. Under the Porter--Thomas and scrambled-noise assumptions, this is numerically compatible with IBM's fidelity lower bound; separately, fidelity-weighted resource accounting projects a 583-contraction workload with a 10.6-minute makespan on the same 32 nodes. More broadly, the approach provides a practical diagnostic for experimental outputs and a quantitative tool for designing future doped Clifford sampling experiments.