Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Duan, Haotong, Chen, Zhongming, Wong, Ngai
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2603.12026
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914387814514688
author Duan, Haotong
Chen, Zhongming
Wong, Ngai
author_facet Duan, Haotong
Chen, Zhongming
Wong, Ngai
contents Tensor networks, which are originally developed for characterizing complex quantum many-body systems, have recently emerged as a powerful framework for capturing high-dimensional probability distributions with strong physical interpretability. This paper systematically studies matrix product states (MPS) for generative modeling and shows that unitary MPS, which is a tensor-network architecture that is both simple and expressive, offers clear benefits for unsupervised learning by reducing ambiguity in parameter updates and improving efficiency. To overcome the inefficiency of standard gradient-based MPS training, we develop a Riemannian optimization approach that casts probabilistic modeling as an optimization problem with manifold constraints, and further derive an efficient space-decoupling algorithm. Experiments on Bars-and-Stripes and EMNIST datasets demonstrate fast adaptation to data structure, stable updates, and strong performance while maintaining the efficiency and expressive power of MPS.
format Preprint
id arxiv_https___arxiv_org_abs_2603_12026
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Efficient Generative Modeling with Unitary Matrix Product States Using Riemannian Optimization
Duan, Haotong
Chen, Zhongming
Wong, Ngai
Machine Learning
Tensor networks, which are originally developed for characterizing complex quantum many-body systems, have recently emerged as a powerful framework for capturing high-dimensional probability distributions with strong physical interpretability. This paper systematically studies matrix product states (MPS) for generative modeling and shows that unitary MPS, which is a tensor-network architecture that is both simple and expressive, offers clear benefits for unsupervised learning by reducing ambiguity in parameter updates and improving efficiency. To overcome the inefficiency of standard gradient-based MPS training, we develop a Riemannian optimization approach that casts probabilistic modeling as an optimization problem with manifold constraints, and further derive an efficient space-decoupling algorithm. Experiments on Bars-and-Stripes and EMNIST datasets demonstrate fast adaptation to data structure, stable updates, and strong performance while maintaining the efficiency and expressive power of MPS.
title Efficient Generative Modeling with Unitary Matrix Product States Using Riemannian Optimization
topic Machine Learning
url https://arxiv.org/abs/2603.12026