Metraipolis › archive › dc_e31bb7cc1ce3
Deep neural networks as lattice gauge theories — arXiv:2608.19331
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.19331
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T11:48:34.693Z
- SHA-256 of the archived text
6b3b69390364ca0dd1e62d4bafe0195d3733040ddd944602f2c6e2d613f4d2de- Stable URL of this page
https://metraipolis.com/documento/dc_e31bb7cc1ce3
How to cite
Deep neural networks as lattice gauge theories — arXiv:2608.19331 — arXiv (en), https://arxiv.org/abs/2608.19331 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e31bb7cc1ce3 (Metraipolis archive dc_e31bb7cc1ce3, sha256 6b3b69390364ca0dd1e62d4bafe0195d3733040ddd944602f2c6e2d613f4d2de)
BibTeX
@misc{dce31bb7cc1ce3,
title = {Deep neural networks as lattice gauge theories — arXiv:2608.19331},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.19331},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e31bb7cc1ce3; sha256 6b3b69390364ca0dd1e62d4bafe0195d3733040ddd944602f2c6e2d613f4d2de},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.19331. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
We modify the NN/QFT duality [1] to incorporate the layerwise permutation symmetry of the network, resulting in a $(0\!+\!1)$-dimensional lattice gauge theory, in which each layer of $N$ neurons acts as an $N$-component lattice site, and the weight matrices play the role of gauge fields living on the links. In this framework, we compute the tree-level neuron-neuron propagator which describes the evolution of layer variance in the network, and develop the Feynman diagram machinery to compute interactions in the perturbative expansion in $1/N$. In particular, we obtain a recursive expression for all corrections to the exact propagator at $O(1)$, representing statistical fluctuations in the ensemble of networks, including infinitely-many loop diagrams mediating the interactions from previous layers. We also present a preliminary analysis of neuron scattering amplitudes that contribute order-by-order in $1/N$, which provides a field-theoretic framework for studying higher-point correlations, and by extension information propagation, in deep networks. We remark on some interesting directions for future work at the intersection of neural networks and quantum field theory. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-21 in cs.LG). Categorie arXiv: High Energy Physics - Theory (hep-th) ; Disordered Systems and Neural Networks (cond-mat.dis-nn); Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2608.19331 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-18T11:47:43.356Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-21?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).