Learning tensor networks with tensor cross interpolation: new algorithms and libraries
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 …
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 …
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 …
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
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 …
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 …
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
Emerging tensor network techniques for solutions of partial differential equations (PDEs),
known for their ability to break the curse of dimensionality, deliver new mathematical …
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
Surrogate models can reduce computational costs for multivariable functions with an
unknown internal structure (black boxes). In a discrete formulation, surrogate modeling is …
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 …
simulation. In particular, such methods are often used in the design and modelling of …
tntorch: Tensor network learning with PyTorch
We present tntorch, a tensor learning framework that supports multiple decompositions
(including Candecomp/Parafac, Tucker, and Tensor Train) under a unified interface. With …
(including Candecomp/Parafac, Tucker, and Tensor Train) under a unified interface. With …
Tensor train for global optimization problems in robotics
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 …
initial guess given to the solver. To address this issue, we propose a novel approach that …