arXiv · 2611.18848
Flexible ground-state search algorithm on hybrid Clifford-tensor network manifolds
Abstract
Tensor network algorithms face challenges in highly entangled regimes, such as systems with periodic boundary conditions (PBC), long-range interactions, or higher dimensions, where the required bond dimension quickly renders computations infeasible. To overcome this limitation, we introduce a novel approach that integrates the stabilizer formalism directly into the traditional density-matrix renormalization group (DMRG) algorithm. While previous methods have used Clifford unitaries to minimize ground-state entanglement, fundamentally limited to distilling that away utilizing local two-qubit operations, our framework optimizes Clifford transformations to minimize the state's energy. We benchmark our method against standard DMRG at fixed bond dimensions for the Ising model on a ring (including long-range interactions) and on a torus. We observe systematic gains in models with local interactions - as drastic as a factor of 40 decrease in bond dimension in two dimensions. Our algorithm is readily implementable in standard DMRG codes and provides a practical way to substantially reduce algorithm costs, paving the way for large-scale simulation at volumes beyond what is presently achievable.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Hari Timsina, Antonio Francesco Mello, Meghadeepa Adhikary, Marcello Dalmonte, Mario Collura. 2026-11-13. Flexible ground-state search algorithm on hybrid Clifford-tensor network manifolds. https://searcharxiv.com/abs/2611.18848
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Searcharxiv by Indiciae: BibTeX · RIS.