analiza aiqbench
This theoretical work bridges neural network ensembles with nuclear reaction theory, revealing that standard models effectively operate in a "closed" regime where uncertainty is purely epistemic. The analysis demonstrates that mainstream deep learning architectures lack the continuous spectrum and wave-like dynamics required to access the "open" case, resulting in a structural inability to capture irreversible probability flux. Consequently, the study maps the precise boundary where current training objectives fail to represent genuine openness, offering a rigorous framework for understanding the limits of reduced-basis emulators in feature-rich regimes.
Integrating Out, Twice:The Open-System Case That Neural-Network Ensemble Theory Is Missing
arXiv:2606.09950v1 Announce Type: new Abstract: Averaging a neural network over its random parameters and marginalizing a Gaussian sector are the same operation, the Schur complement of the eliminated block, and when that block is closed it returns a covariance and its inverse. That is all a network ensemble produces, the closed case. The open case is missing, and nuclear reaction theory has it worked out. Projecting a scattering problem onto a chosen set of channels, with the rest carrying probability irreversibly to a continuum, leaves a non-Hermitian effective generator that conserves and itemizes exactly what it loses: the nuclear optical model and its generalized optical theorem. I set the two cases side by side using only the moments of a distribution, the algebra of Gaussians, and block inversion, no field theory, and give the closed-case dictionary in full: the neural tangent kernel is the Fisher sensitivity kernel, the infinite-width Gaussian limit is the Gaussian-process emulator, and the lazy-to-feature transition is the validity boundary of a reduced-basis emulator. I then test the open export on a truncated attention map, a token-level transfer operator, and a sparse expert router, and report a mostly negative result. The conserved flux ledger ports wherever openness is genuinely present, but its distinctive content is absent, an artifact of the chosen partition, or pinned near a floor by the training objective, and the operationally useful uncertainty turns out to be epistemic, living in the closed half of the correspondence, not the open one. The negative has a structural reason this note makes precise: the open case needs an eliminated sector with a continuous spectrum and wave-like, not relaxational, dynamics, which mainstream learning's finite or dissipative objects do not supply. This is a note, not a result; its main finding is that negative one, and its value is the map that locates it.
Abstract:Averaging a neural network over its random parameters and marginalizing a Gaussian sector are the same operation, the Schur complement of the eliminated block, and when that block is closed it returns a covariance and its inverse. That is all a network ensemble produces, the closed case. The open case is missing, and nuclear reaction theory has it worked out. Projecting a scattering problem onto a chosen set of channels, with the rest carrying probability irreversibly to a continuum, leaves a non-Hermitian effective generator that conserves and itemizes exactly what it loses: the nuclear optical model and its generalized optical theorem. I set the two cases side by side using only the moments of a distribution, the algebra of Gaussians, and block inversion, no field theory, and give the closed-case dictionary in full: the neural tangent kernel is the Fisher sensitivity kernel, the infinite-width Gaussian limit is the Gaussian-process emulator, and the lazy-to-feature transition is the validity boundary of a reduced-basis emulator. I then test the open export on a truncated attention map, a token-level transfer operator, and a sparse expert router, and report a mostly negative result. The conserved flux ledger ports wherever openness is genuinely present, but its distinctive content is absent, an artifact of the chosen partition, or pinned near a floor by the training objective, and the operationally useful uncertainty turns out to be epistemic, living in the closed half of the correspondence, not the open one. The negative has a structural reason this note makes precise: the open case needs an eliminated sector with a continuous spectrum and wave-like, not relaxational, dynamics, which mainstream learning's finite or dissipative objects do not supply. This is a note, not a result; its main finding is that negative one, and its value is the map that locates it.
Machine Learning (cs.LG); Nuclear Theory (nucl-th); Computational Physics (physics.comp-ph); Data Analysis, Statistics and Probability (physics.data-an)
arXiv:2606.09950 [cs.LG]
(or arXiv:2606.09950v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2606.09950 arXiv-issued DOI via DataCite (pending registration)
Text integral sub licența deschisă a publicației · 330 cuvinte
Subiecte conexe
NVIDIA GTC Taipei at COMPUTEX: Live Updates on What’s Next in AI
NVIDIA · 21 mai 2026
Humanity’s Last Invention — Richard Socher of Recursive
Latent Space · 14 sept. 2026
[AINews] High Taste Testers Ep. 1
Latent Space · 17 oct. 2026
How Photonic Chips Differ From Electronic Chips
Quantum Zeitgeist · 19 sept. 2026