Enregistré dans:
Détails bibliographiques
Auteurs principaux: Zhao, Hanyu, Wu, Yang, Hou, Yuexian
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:https://arxiv.org/abs/2603.04451
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866915835036041216
author Zhao, Hanyu
Wu, Yang
Hou, Yuexian
author_facet Zhao, Hanyu
Wu, Yang
Hou, Yuexian
contents Inspired by measurement incompatibility and Bell-family inequalities in quantum mechanics, we propose the Non-Classical Network (NCnet), a simple classical neural architecture that stably exhibits non-classical statistical behaviors under typical and interpretable experimental setups. We find non-classicality, measured by the $S$ statistic of CHSH inequality, arises from gradient competitions of hidden-layer neurons shared by multi-tasks. Remarkably, even without physical links supporting explicit communication, one task head can implicitly sense the training task of other task heads via local loss oscillations, leading to non-local correlations in their training outcomes. Specifically, in the low-resource regime, the value of $S$ increases gradually with increasing resources and approaches toward its classical upper-bound 2, which implies that underfitting is alleviated with resources increase. As the model nears the critical scale required for adequate performance, $S$ may temporarily exceed 2. As resources continue to grow, $S$ then asymptotically decays down to and fluctuates around 2. Empirically, when model capacity is insufficient, $S$ is positively correlated with generalization performance, and the regime where $S$ first approaches $2$ often corresponding to good generalization. Overall, our results suggest that non-classical statistics can provide a novel perspective for understanding internal interactions and training dynamics of deep networks.
format Preprint
id arxiv_https___arxiv_org_abs_2603_04451
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle On Emergences of Non-Classical Statistical Characteristics in Classical Neural Networks
Zhao, Hanyu
Wu, Yang
Hou, Yuexian
Machine Learning
Artificial Intelligence
Quantum Physics
Inspired by measurement incompatibility and Bell-family inequalities in quantum mechanics, we propose the Non-Classical Network (NCnet), a simple classical neural architecture that stably exhibits non-classical statistical behaviors under typical and interpretable experimental setups. We find non-classicality, measured by the $S$ statistic of CHSH inequality, arises from gradient competitions of hidden-layer neurons shared by multi-tasks. Remarkably, even without physical links supporting explicit communication, one task head can implicitly sense the training task of other task heads via local loss oscillations, leading to non-local correlations in their training outcomes. Specifically, in the low-resource regime, the value of $S$ increases gradually with increasing resources and approaches toward its classical upper-bound 2, which implies that underfitting is alleviated with resources increase. As the model nears the critical scale required for adequate performance, $S$ may temporarily exceed 2. As resources continue to grow, $S$ then asymptotically decays down to and fluctuates around 2. Empirically, when model capacity is insufficient, $S$ is positively correlated with generalization performance, and the regime where $S$ first approaches $2$ often corresponding to good generalization. Overall, our results suggest that non-classical statistics can provide a novel perspective for understanding internal interactions and training dynamics of deep networks.
title On Emergences of Non-Classical Statistical Characteristics in Classical Neural Networks
topic Machine Learning
Artificial Intelligence
Quantum Physics
url https://arxiv.org/abs/2603.04451