Transforms empirical loss into Gibbs measures on hierarchical structures to characterize families of equilibrium learning states via fixed-point equations and phase transitions on Cayley trees.
Ariosto,Statistical Physics of Deep Neural Networks: Generalization Capability, Beyond the Infinite Width, and Feature Learning, 2025, 10.48550/arXiv.2501.19281
1 Pith paper cite this work, alongside 1 external citations. Polarity classification is still indexing.
1
Pith paper citing it
1
external citations · external index
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Data-Driven Energy-Based Learning via Gibbs Measures on Hierarchical Structures
Transforms empirical loss into Gibbs measures on hierarchical structures to characterize families of equilibrium learning states via fixed-point equations and phase transitions on Cayley trees.