Copertina

Universal characteristics of deep neural network loss surfaces from random matrix theory

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Baskerville, Nicholas P, Keating, Jonathan P, Mezzadri, Francesco, Najnudel, Joseph, Granziol, Diego
Natura:	Preprint
Pubblicazione:	2022
Soggetti:	Mathematical Physics Disordered Systems and Neural Networks Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

A spin-glass model for the loss surfaces of generative adversarial networks
di: Baskerville, Nicholas P, et al.
Pubblicazione: (2021)

The thermodynamic limit in mean field neural networks
di: Agliari, Elena, et al.
Pubblicazione: (2024)

A spectral approach to Hebbian-like neural networks
di: Agliari, Elena, et al.
Pubblicazione: (2024)

Serial vs parallel recall in the Blume-Every-Griffiths neural networks
di: Albanese, Linda, et al.
Pubblicazione: (2026)

Renormalization group for deep neural networks: Universality of learning and scaling laws
di: Coppola, Gorka Peraza, et al.
Pubblicazione: (2025)

Improving deep neural network performance through sampling
di: Ghantasala, Lakshmi A., et al.
Pubblicazione: (2025)

The importance of being empty: a spectral approach to Hopfield neural networks with diluted examples
di: Agliari, Elena, et al.
Pubblicazione: (2025)

Critical feature learning in deep neural networks
di: Fischer, Kirsten, et al.
Pubblicazione: (2024)

Networks of neural networks: more is different
di: Agliari, Elena, et al.
Pubblicazione: (2025)

The Loss Surfaces of Neural Networks with General Activation Functions
di: Baskerville, Nicholas P., et al.
Pubblicazione: (2020)

A statistical mechanics framework for Bayesian deep neural networks beyond the infinite-width limit
di: Pacelli, R., et al.
Pubblicazione: (2022)

On the asymptotic duality of spectral variances in random matrix theory and the "1/6" formula
di: Tian, Peng, et al.
Pubblicazione: (2026)

Towards a theory of how the structure of language is acquired by deep neural networks
di: Cagnetta, Francesco, et al.
Pubblicazione: (2024)

Information-theoretic reduction of deep neural networks to linear models in the overparametrized proportional regime
di: Camilli, Francesco, et al.
Pubblicazione: (2025)

Arbitrage equilibrium and the emergence of universal microstructure in deep neural networks
di: Venkatasubramanian, Venkat, et al.
Pubblicazione: (2024)

Transfer matrix analysis of non-hermitian Hamiltonians: asymptotic spectra and topological eigenvalues
di: Koekenbier, Lars, et al.
Pubblicazione: (2024)

Concentration properties of Gaussian random fields
di: Sharma, Gargee
Pubblicazione: (2014)

Bounds in partition functions of the continuous random field Ising model
di: Heymans, G. O., et al.
Pubblicazione: (2024)

Gauge theory for quantum XYZ spin glasses
di: Itoi, C., et al.
Pubblicazione: (2023)

The weak and strong disorder regimes in the continuous random field Ising model
di: Heymans, G. O., et al.
Pubblicazione: (2026)

Disordered harmonic chain with random masses and springs: a combinatorial approach
di: Bernard, Maximilien, et al.
Pubblicazione: (2025)

Higher-order localization landscape theory of Anderson localization
di: Skipetrov, Sergey E.
Pubblicazione: (2024)

Implicit bias produces neural scaling laws in learning curves, from perceptrons to deep networks
di: D'Amico, Francesco, et al.
Pubblicazione: (2025)

Analyzing partially-polarized light with a photonic deep random neural network
di: Petrini, Alessandro, et al.
Pubblicazione: (2025)

Dynamical fluctuations of random walks in higher-order networks
di: Di Gaetano, Leonardo, et al.
Pubblicazione: (2023)

On the synaptic matrix eigenvalues of sparsely connected neural networks
di: Ansari, Mohd. Gayas, et al.
Pubblicazione: (2026)

Effective medium theory for the electrical conductivity of random metallic nanowire networks
di: Tarasevich, Yuri Yu., et al.
Pubblicazione: (2025)

Emergent weight morphologies in deep neural networks
di: de Jong, Pascal, et al.
Pubblicazione: (2025)

Winding Number Statistics for Chiral Random Matrices: Universal Correlations and Statistical Moments in the Unitary Case
di: Hahn, Nico, et al.
Pubblicazione: (2024)

Spring-block theory of feature learning in deep neural networks
di: Shi, Cheng, et al.
Pubblicazione: (2024)

Large deviations of spectral determinants of matrix-valued random Schrödinger operators and Dyson Brownian motion in cubic potentials
di: Fyodorov, Yan, et al.
Pubblicazione: (2025)

Kac-Rice inspired approach to non-Hermitian random matrices
di: Fyodorov, Yan V
Pubblicazione: (2025)

Computing frustration and near-monotonicity in deep neural networks
di: Wendin, Joel, et al.
Pubblicazione: (2025)

Finite-time Lyapunov exponents of deep neural networks
di: Storm, L., et al.
Pubblicazione: (2023)

Average-case matrix discrepancy: satisfiability bounds
di: Maillard, Antoine
Pubblicazione: (2024)

Application of deep neural networks for computing the renormalization group flow of the two-dimensional phi^4 field theory
di: Zhao, Yueqi, et al.
Pubblicazione: (2025)

Spectral boundaries of deterministic matrices deformed by rotationally invariant random non-Hermitian ensembles
di: Bousseyroux, Pierre, et al.
Pubblicazione: (2026)

Transmission eigenvalue distribution in disordered media from radiant field theory
di: Gaspard, David, et al.
Pubblicazione: (2024)

Anomalous random networks
di: Zhang, Hong, et al.
Pubblicazione: (2024)

Expansion creates spin-glass order in finite-connectivity models: a rigorous and intuitive approach from the theory of LDPC codes
di: Placke, Benedikt, et al.
Pubblicazione: (2025)