
















Abstract:Neural Network (NN) architectures that break statistical independence of parameters have been proposed as a new approach for simulating local quantum field theories (QFTs). In the infinite neuron number limit, single-layer NNs can exactly reproduce QFT results. This paper examines the viability of this architecture for perturbative calculations of local QFTs for finite neuron number $N$ using scalar $\phi^4$ theory in $d$ Euclidean dimensions as an example. We find that the renormalized $O(1/N)$ corrections to two- and four-point correlators yield perturbative series which are sensitive to the ultraviolet cut-off and therefore have a weak convergence. We propose a modification to the architecture to improve this convergence and discuss constraints on the parameters of the theory and the scaling of N which allow us to extract accurate field theory results.
| Comments: | Published version |
| Subjects: | High Energy Physics - Theory (hep-th); Machine Learning (cs.LG) |
| Cite as: | arXiv:2508.03810 [hep-th] |
| (or arXiv:2508.03810v4 [hep-th] for this version) | |
| https://doi.org/10.48550/arXiv.2508.03810 arXiv-issued DOI via DataCite |
From: Srimoyee Sen [view email]
[v1]
Tue, 5 Aug 2025 18:00:31 UTC (365 KB)
[v2]
Sat, 6 Sep 2025 18:37:48 UTC (365 KB)
[v3]
Tue, 21 Oct 2025 16:47:43 UTC (365 KB)
[v4]
Thu, 14 May 2026 20:07:53 UTC (241 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理,仅供阅读参考。 原文来自 — 版权归原作者所有。