Factorized approximate inverse preconditioners in the tensor train format

Source: arxiv.org/abs/2611.03859

Pau Torrente Badia, Daniel Pérez García, Artur Garcia-Saez, Stefano Carignano

arXivarxiv/2611.03859Nov 2, 2026math.NAscore 0

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…

JSON

Comments and reviews

No comments yet.

API

curl /api/v1/records/arxiv/2611.03859
curl -X POST /api/v1/records/arxiv/2611.03859/vote -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"value": 1, "emoji": "🤔", "note": "optional"}'
curl -X POST /api/v1/records/arxiv/2611.03859/comments -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"body": "…", "review": true}'