Saved in:
Bibliographic Details
Main Authors: Shokry, Ahmedeo, Santini, Alessandro, Vicentini, Filippo
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.08430
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911248512188416
author Shokry, Ahmedeo
Santini, Alessandro
Vicentini, Filippo
author_facet Shokry, Ahmedeo
Santini, Alessandro
Vicentini, Filippo
contents The natural gradient is central in neural quantum states optimizations but it is limited by the cost of computing and inverting the quantum geometric tensor, the quantum analogue of the Fisher information matrix. We introduce a block-diagonal quantum geometric tensor that partitions the metric by network layers, analogous to block-structured Fisher methods such as K-FAC. This layer-wise approximation preserves essential curvature while removing noisy cross-layer correlations, improving conditioning and scalability. Experiments on Heisenberg and frustrated $J_1$-$J_2$ models show faster convergence, lower energy, and improved stability.
format Preprint
id arxiv_https___arxiv_org_abs_2510_08430
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle When Less is More: Approximating the Quantum Geometric Tensor with Block Structures
Shokry, Ahmedeo
Santini, Alessandro
Vicentini, Filippo
Quantum Physics
Disordered Systems and Neural Networks
Computational Physics
The natural gradient is central in neural quantum states optimizations but it is limited by the cost of computing and inverting the quantum geometric tensor, the quantum analogue of the Fisher information matrix. We introduce a block-diagonal quantum geometric tensor that partitions the metric by network layers, analogous to block-structured Fisher methods such as K-FAC. This layer-wise approximation preserves essential curvature while removing noisy cross-layer correlations, improving conditioning and scalability. Experiments on Heisenberg and frustrated $J_1$-$J_2$ models show faster convergence, lower energy, and improved stability.
title When Less is More: Approximating the Quantum Geometric Tensor with Block Structures
topic Quantum Physics
Disordered Systems and Neural Networks
Computational Physics
url https://arxiv.org/abs/2510.08430