Communication Dans Un Congrès Année : 2025

Near-Optimal Contraction Strategies for the Scalar Product in the Tensor-Train Format

Résumé

Tensor-train (TT) decomposition has garnered tremendous popularity for its efficiency in handling high-dimensional data arising in scientific and quantum computing as well as machine learning applications. It provides a compact representation for matrices and vectors with a Kronecker product-like low-rank structure and enables efficient matrix-vector operations in this compressed form. The vector scalar product is among such key operations, comprising a series of tensor contractions in a specific tensor network topology whose order significantly impacts the computational cost. In this work, we propose efficient algorithms for finding near-optimal contraction orderings for tensor networks representing scalar products in the TT format. We show that our algorithms outperform all existing contraction ordering methods for general tensor networks where the best existing method incurs up to 15% higher cost for x T y, twice the cost for x T Ay, and ten times higher cost for x T ABy scalar products where x, y and A, B are vectors and matrices expressed in the TT format, respectively.

Fichier principal
Vignette du fichier
_EuroPar25__Exploring_optimal_contraction_strategies_for_fast_vector_scalar_product_in_Tensor_Train_format-HAL.pdf (829.89 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-05285400 , version 1 (26-09-2025)

Licence

Identifiants

Citer

Atte Torri, Przemysław Dominikowski, Brice Pointal, Oguz Kaya, Laércio Lima Pilla, et al.. Near-Optimal Contraction Strategies for the Scalar Product in the Tensor-Train Format. Euro-Par 2025 - 31 International European Conference on Parallel and Distributed Computing, Aug 2025, Dresden, Germany. pp.63-77, ⟨10.1007/978-3-031-99872-0_5⟩. ⟨hal-05285400⟩
2688 Consultations
238 Téléchargements

Altmetric

Partager

  • More