Learning tensor networks with tensor cross interpolation: new algorithms and libraries

YN Fernández, MK Ritter, M Jeannin, JW Li… - arXiv preprint arXiv …, 2024 - arxiv.org
The tensor cross interpolation (TCI) algorithm is a rank-revealing algorithm for decomposing
low-rank, high-dimensional tensors into tensor trains/matrix product states (MPS). TCI learns …

Tensor Train Optimization for Conformational Sampling of Organic Molecules

C Zurek, RA Mallaev, AC Paul… - Journal of Chemical …, 2024 - ACS Publications
Exploring the conformational space of molecules remains a challenge of fundamental
importance to quantum chemistry: identification of relevant conformers at ambient conditions …

PROTES: probabilistic optimization with tensor sampling

A Batsheva, A Chertkov… - Advances in Neural …, 2023 - proceedings.neurips.cc
We developed a new method PROTES for black-box optimization, which is based on the
probabilistic sampling from a probability density function given in the low-parametric tensor …

Alternating local enumeration (tnale): Solving tensor network structure search with fewer evaluations

C Li, J Zeng, C Li, CF Caiafa… - … Conference on Machine …, 2023 - proceedings.mlr.press
Tensor network (TN) is a powerful framework in machine learning, but selecting a good TN
model, known as TN structure search (TN-SS), is a challenging and computationally …

Towards practical control of singular values of convolutional layers

A Senderovich, E Bulatova… - Advances in Neural …, 2022 - proceedings.neurips.cc
In general, convolutional neural networks (CNNs) are easy to train, but their essential
properties, such as generalization error and adversarial robustness, are hard to control …

[HTML][HTML] Tensor Network Space-Time Spectral Collocation Method for Time-Dependent Convection-Diffusion-Reaction Equations

D Adak, DP Truong, G Manzini, KØ Rasmussen… - Mathematics, 2024 - mdpi.com
Emerging tensor network techniques for solutions of partial differential equations (PDEs),
known for their ability to break the curse of dimensionality, deliver new mathematical …

Black box approximation in the tensor train format initialized by ANOVA decomposition

A Chertkov, G Ryzhakov, I Oseledets - SIAM Journal on Scientific Computing, 2023 - SIAM
Surrogate models can reduce computational costs for multivariable functions with an
unknown internal structure (black boxes). In a discrete formulation, surrogate modeling is …

Protein-protein docking using a tensor train black-box optimization method

D Morozov, A Melnikov, V Shete… - arXiv preprint arXiv …, 2023 - arxiv.org
Black-box optimization methods play an important role in many fields of computational
simulation. In particular, such methods are often used in the design and modelling of …

tntorch: Tensor network learning with PyTorch

M Usvyatsov, R Ballester-Ripoll, K Schindler - Journal of Machine Learning …, 2022 - jmlr.org
We present tntorch, a tensor learning framework that supports multiple decompositions
(including Candecomp/Parafac, Tucker, and Tensor Train) under a unified interface. With …

Tensor train for global optimization problems in robotics

S Shetty, T Lembono, T Loew… - … International Journal of …, 2024 - journals.sagepub.com
The convergence of many numerical optimization techniques is highly dependent on the
initial guess given to the solver. To address this issue, we propose a novel approach that …