Generative Modeling via Hierarchical Tensor Sketching
Numerical Analysis
2026-01-13 v2 Machine Learning
Numerical Analysis
Machine Learning
Abstract
We propose a hierarchical tensor-network approach for approximating high-dimensional probability density via empirical distribution. This leverages randomized singular value decomposition (SVD) techniques and involves solving linear equations for tensor cores in this tensor network. The complexity of the resulting algorithm scales linearly in the dimension of the high-dimensional density. An analysis of estimation error demonstrates the effectiveness of this method through several numerical experiments.
Cite
@article{arxiv.2304.05305,
title = {Generative Modeling via Hierarchical Tensor Sketching},
author = {Yifan Peng and Yian Chen and E. Miles Stoudenmire and Yuehaw Khoo},
journal= {arXiv preprint arXiv:2304.05305},
year = {2026}
}
Comments
31 pages, 15 figures