Copertina

Stochastic Scaling Limits and Synchronization by Noise in Deep Transformer Models

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Agazzi, Andrea, Bruno, Giuseppe, García, Eloy Mosig, Saviozzi, Samuele, Romito, Marco
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Probability Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Quantitative convergence of trained single layer neural networks to Gaussian processes
di: Mosig, Eloy, et al.
Pubblicazione: (2025)

Global Optimization via Softmin Energy Minimization
di: Agazzi, Andrea, et al.
Pubblicazione: (2025)

A multiscale analysis of mean-field transformers in the moderate interaction regime
di: Bruno, Giuseppe, et al.
Pubblicazione: (2025)

Universality in Deep Neural Networks: An approach via the Lindeberg exchange principle
di: Giovagnini, Filippo, et al.
Pubblicazione: (2026)

Scaling Limits of Long-Context Transformers
di: Bruno, Giuseppe, et al.
Pubblicazione: (2026)

Effective continuous equations for adaptive SGD: a stochastic analysis view
di: Callisti, Luca, et al.
Pubblicazione: (2025)

Uniform Scaling Limits in AdamW-Trained Transformers
di: Gibson, William, et al.
Pubblicazione: (2026)

Random Quadratic Form on a Sphere: Synchronization by Common Noise
di: Engel, Maximilian, et al.
Pubblicazione: (2026)

Limit Theorems for Stochastic Gradient Descent with Infinite Variance
di: Blanchet, Jose, et al.
Pubblicazione: (2024)

Decoupled Functional Central Limit Theorems for Two-Time-Scale Stochastic Approximation
di: Han, Yuze, et al.
Pubblicazione: (2024)

Variational Kernel Design for Internal Noise: Gaussian Chaos Noise, Representation Compatibility, and Reliable Deep Learning
di: Liu, Ziran
Pubblicazione: (2026)

Noise-induced stabilization in a chemical reaction network without boundary effects
di: Agazzi, Andrea, et al.
Pubblicazione: (2025)

Matrix Denoising with Doubly Heteroscedastic Noise: Fundamental Limits and Optimal Spectral Methods
di: Zhang, Yihan, et al.
Pubblicazione: (2024)

Partially Stochastic Infinitely Deep Bayesian Neural Networks
di: Calvo-Ordonez, Sergio, et al.
Pubblicazione: (2024)

Approximation to Deep Q-Network by Stochastic Delay Differential Equations
di: Lu, Jianya, et al.
Pubblicazione: (2025)

Efficiency of Parallel and Restart Exploration Strategies in Model Free Stochastic Simulations
di: Garcia, Ernesto, et al.
Pubblicazione: (2025)

Concentration of General Stochastic Approximation Under Heavy-Tailed Markovian Noise
di: Agrawal, Shubhada, et al.
Pubblicazione: (2026)

Algorithmic Stability of Stochastic Gradient Descent with Momentum under Heavy-Tailed Noise
di: Dang, Thanh, et al.
Pubblicazione: (2025)

On the Limits of Latent Reuse in Diffusion Models
di: Yu, Yifeng, et al.
Pubblicazione: (2026)

Random-Bridges as Stochastic Transports for Generative Models
di: Goria, Stefano, et al.
Pubblicazione: (2025)

Limit Theorems for Stochastic Gradient Descent in High-Dimensional Single-Layer Networks
di: Rangriz, Parsa
Pubblicazione: (2025)

Flatness-Aware Stochastic Gradient Langevin Dynamics
di: Bruno, Stefano, et al.
Pubblicazione: (2025)

Information-Theoretic Limits and Strong Consistency on Binary Non-uniform Hypergraph Stochastic Block Models
di: Wang, Hai-Xiao
Pubblicazione: (2023)

Stochastic Differential Equations models for Least-Squares Stochastic Gradient Descent
di: Schertzer, Adrien, et al.
Pubblicazione: (2024)

Mixed Noise and Posterior Estimation with Conditional DeepGEM
di: Hagemann, Paul, et al.
Pubblicazione: (2024)

Stochastic Operator Network: A Stochastic Maximum Principle Based Approach to Operator Learning
di: Bausback, Ryan, et al.
Pubblicazione: (2025)

Scaling Laws from Sequential Feature Recovery: A Solvable Hierarchical Model
di: Wortsman-Zurich, Arie, et al.
Pubblicazione: (2026)

Some Theoretical Limitations of t-SNE
di: Li, Rupert, et al.
Pubblicazione: (2026)

Decentralized Proximal Stochastic Gradient Langevin Dynamics
di: Islam, Mohammad Rafiqul, et al.
Pubblicazione: (2026)

Adaptive Learning via Off-Model Training and Importance Sampling for Fully Non-Markovian Optimal Stochastic Control. Complete version
di: Leão, Dorival, et al.
Pubblicazione: (2026)

Stabilizing Temporal Difference Learning via Implicit Stochastic Recursion
di: Kim, Hwanwoo, et al.
Pubblicazione: (2025)

Flow Matching: Markov Kernels, Stochastic Processes and Transport Plans
di: Wald, Christian, et al.
Pubblicazione: (2025)

Convergence of Two Time-Scale Stochastic Approximation: A Martingale Approach
di: Vidyasagar, Mathukumalli
Pubblicazione: (2026)

Homogenized Transformers
di: Koubbi, Hugo, et al.
Pubblicazione: (2026)

Gaussian Fluctuations for the Stochastic Landau-Lifshitz Navier-Stokes Equation in Dimension $D\geq2$
di: Kotitsas, Sotiris, et al.
Pubblicazione: (2025)

Optimal Initialization in Depth: Lyapunov Initialization and Limit Theorems for Deep Leaky ReLU Networks
di: Kogler, Constantin, et al.
Pubblicazione: (2026)

Stochastic Port-Hamiltonian Neural Networks: Universal Approximation with Passivity Guarantees
di: Di Persio, Luca, et al.
Pubblicazione: (2026)

Non-Asymptotic Convergence of Stochastic Iterative Algorithms: A Lyapunov Framework
di: Chen, Zaiwei, et al.
Pubblicazione: (2026)

Convergence, Sticking and Escape: Stochastic Dynamics Near Critical Points in SGD
di: Dudukalov, Dmitry, et al.
Pubblicazione: (2025)

Exact Gradients for Stochastic Spiking Neural Networks Driven by Rough Signals
di: Holberg, Christian, et al.
Pubblicazione: (2024)