Factorized approximate inverse preconditioners in the tensor train format

arXiv:2611.03859 · math.NA, cs.NA, quant-ph · Submitted 2026-11-02 · Read on arXiv

math.NA, cs.NA, quant-ph

Submitted: 2026-11-02

Updated: 2026-11-02

Comments: 50 pages, 11 figures

Code: https://github.com/bsc-quantic/DifferenTT.jl

License: http://creativecommons.org/licenses/by/4.0/

The gist: We present a preconditioning pipeline that accelerates Tensor Train (TT) linear solvers for symmetric positive-definite problems.

Terminology

Abstract

We present a preconditioning pipeline that accelerates Tensor Train (TT) linear solvers for symmetric positive-definite problems. Extending the Factorized Sparse Approximate Inverse framework to tensor networks, we construct a preconditioner as the product of a TT operator and its transpose. This formulation is positive semidefinite for any bond dimension and truncation, and definite whenever the factor is non-singular, satisfying the mathematical requirements of the Conjugate Gradient method. We detail the construction of this preconditioner and its integration into Density Matrix Renormalization Group (DMRG)-like linear solvers. Performance evaluations show consistent iteration reductions for anisotropic diffusion and Poisson problems across various configurations, leading to net speedups of up to 25% in ill-conditioned settings of time stepping schemes that can reuse the approximate inverses. These benchmarks were performed with the open-source Julia library DifferenTT.jl, which we also introduce in this work to provide out-of-the-box access to these preconditioned pipelines.

Sources

Related papers