The neural networks with tensor weights and emergent fermionic Wick rules in the large-width limit

Abstract

In this paper, we study complex-valued neural network (CVNNs) with tensor-valued hidden-to-output weights within the framework of neural-network quantum field theory (NN-QFT). For standard CVNNs with scalar weights, we derive the generating functional and identify the exact Gaussian process that arises in the infinite-width limit, together with its associated effective quantum state. When the last-layer weights are promoted to Clifford-algebra-valued tensors, the network output becomes complex matrix-valued, and a fermion-like sign structure in the large-width correlation functions of the network output is induced. We show that, in the infinite-width limit, correlators with equal numbers of $f^{†}$ and $f$ obey fermionic Wick rules and can be written as determinants built from a scalar Euclidean kernel $S(x,y)=⟨f^{†}(x)f(y)⟩$. This provides a sign-structured extension of NN-QFT at the level of Euclidean correlators and Feynman rules, even though a microscopic Grassmann path integral representation for the network parameters has not yet been constructed. Our analysis thus pushes the NN-QFT correspondence beyond purely bosonic Gaussian fields and suggests a possible route to encoding fermion-like symmetries in neural architectures for QFT correspondence.

References in corpus (21)

• Solving the Quantum Many-Body Problem with Artificial Neural Networks
• Discovering Phase Transitions with Unsupervised Learning
• Quantum optics in the phase space - A tutorial on Gaussian states
• Deep Potential: a general representation of a many-body potential energy surface
• Fermionic neural-network states for ab-initio electronic structure
• Solving Statistical Mechanics Using Variational Autoregressive Networks
• Regressive and generative neural networks for scalar field theory
• Discovering Quantum Phase Transitions with Fermionic Neural Networks
• Exploring QCD matter in extreme conditions with Machine Learning
• Neural Networks and Quantum Field Theory
• Complex-valued neural networks for machine learning on non-stationary physical data
• Deep learning stochastic processes with QCD phase transition
• Self-learning Hybrid Monte Carlo: A First-principles Approach
• Quantum field-theoretic machine learning
• Nonperturbative renormalization for the neural network-QFT correspondence
• Diffusion Models as Stochastic Quantization in Lattice Field Theory
• Fourier-Flow model generating Feynman paths
• Unifying view of fermionic neural network quantum states: From neural network backflow to hidden fermion determinant states
• Lecture note on Clifford algebra
• On learning higher-order cumulants in diffusion models
• Interrelation of equivariant Gaussian processes and convolutional neural networks

Cited by in corpus (1)

• Conformal Defects in Neural Network Field Theories