Image de couverture de livre

Factual recall in linear associative memories: sharp asymptotics and mechanistic insights

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Giorlandino, Alessio, Goldt, Sebastian, Maillard, Antoine
Format:	Preprint
Publié:	2026
Sujets:	Machine Learning Disordered Systems and Neural Networks Statistical Mechanics
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Two failure modes of deep transformers and how to avoid them: a unified theory of signal propagation at initialisation
par: Giorlandino, Alessio, et autres
Publié: (2025)

Memorisation, convergence and generalisation in generative models
par: Maillard, Antoine, et autres
Publié: (2026)

A Fourier perspective on the learning dynamics of neural networks: from sample complexities to mechanistic insights
par: Ricci, Fabiola, et autres
Publié: (2026)

Feature learning from non-Gaussian inputs: the case of Independent Component Analysis in high dimensions
par: Ricci, Fabiola, et autres
Publié: (2025)

Mapping of attention mechanisms to a generalized Potts model
par: Rende, Riccardo, et autres
Publié: (2023)

Pseudo-likelihood produces associative memories able to generalize, even for asymmetric couplings
par: D'Amico, Francesco, et autres
Publié: (2025)

Thermodynamics of bidirectional associative memories
par: Barra, Adriano, et autres
Publié: (2022)

The committee machine: Computational to statistical gaps in learning a two-layers neural network
par: Aubin, Benjamin, et autres
Publié: (2018)

Generalization Dynamics of Linear Diffusion Models
par: Merger, Claudia, et autres
Publié: (2025)

On the role of non-linear latent features in bipartite generative neural networks
par: Bonnaire, Tony, et autres
Publié: (2025)

Short-term plasticity recalls forgotten memories through a trampoline mechanism
par: Del Gaudio, Martina, et autres
Publié: (2025)

Weight fluctuations in (deep) linear neural networks and a derivation of the inverse-variance flatness relation
par: Gross, Markus, et autres
Publié: (2023)

Multi-channel pattern reconstruction through $L$-directional associative memories
par: Agliari, Elena, et autres
Publié: (2025)

Semi-analytic approximate stability selection for correlated data in generalized linear models
par: Takahashi, Takashi, et autres
Publié: (2020)

How transformers learn structured data: insights from hierarchical filtering
par: Garnier-Brun, Jerome, et autres
Publié: (2024)

Distinct mechanisms underlying in-context learning in transformers
par: Gibson, Cole, et autres
Publié: (2026)

Interpreting the Synchronization Gap: The Hidden Mechanism Inside Diffusion Transformers
par: Albrychiewicz, Emil, et autres
Publié: (2026)

Escape dynamics and implicit bias of one-pass SGD in overparameterized quadratic networks
par: Bocchi, Dario, et autres
Publié: (2026)

From Classical to Quantum: Extending Prometheus for Unsupervised Discovery of Phase Transitions in Three Dimensions and Quantum Systems
par: Yee, Brandon, et autres
Publié: (2026)

The Symmetric Perceptron: a Teacher-Student Scenario
par: Catania, Giovanni, et autres
Publié: (2026)

Rare Event Analysis of Large Language Models
par: Dorman, Jake McAllister, et autres
Publié: (2026)

How does Chain of Thought decompose complex tasks?
par: Nadgir, Amrut, et autres
Publié: (2026)

Biased Generalization in Diffusion Models
par: Garnier-Brun, Jerome, et autres
Publié: (2026)

Analytic theory of dropout regularization
par: Mori, Francesco, et autres
Publié: (2025)

The Copycat Perceptron: Smashing Barriers Through Collective Learning
par: Catania, Giovanni, et autres
Publié: (2023)

Fundamental operating regimes, hyper-parameter fine-tuning and glassiness: towards an interpretable replica-theory for trained restricted Boltzmann machines
par: Fachechi, Alberto, et autres
Publié: (2024)

The autoregressive neural network architecture of the Boltzmann distribution of pairwise interacting spins systems
par: Biazzo, Indaco
Publié: (2023)

Optimal Protocols for Continual Learning via Statistical Physics and Control Theory
par: Mori, Francesco, et autres
Publié: (2024)

Universal Scaling Laws of Absorbing Phase Transitions in Artificial Deep Neural Networks
par: Tamai, Keiichi, et autres
Publié: (2023)

Inferring Higher-Order Couplings with Neural Networks
par: Decelle, Aurélien, et autres
Publié: (2025)

Explaining the effects of non-convergent sampling in the training of Energy-Based Models
par: Agoritsas, Elisabeth, et autres
Publié: (2023)

Discrete generative diffusion models without stochastic differential equations: a tensor network approach
par: Causer, Luke, et autres
Publié: (2024)

The Physics of Data and Tasks: Theories of Locality and Compositionality in Deep Learning
par: Favero, Alessandro
Publié: (2025)

Sampling with flows, diffusion and autoregressive neural networks: A spin-glass perspective
par: Ghio, Davide, et autres
Publié: (2023)

Stochastic Gradient Descent-like relaxation is equivalent to Metropolis dynamics in discrete optimization and inference problems
par: Angelini, Maria Chiara, et autres
Publié: (2023)

Inferring effective couplings with Restricted Boltzmann Machines
par: Decelle, Aurélien, et autres
Publié: (2023)

Transfer Learning in $\ell_1$ Regularized Regression: Hyperparameter Selection Strategy based on Sharp Asymptotic Analysis
par: Okajima, Koki, et autres
Publié: (2024)

A theoretical framework for overfitting in energy-based modeling
par: Catania, Giovanni, et autres
Publié: (2025)

Uncertainty in AI-driven Monte Carlo simulations
par: Tzivrailis, Dimitrios, et autres
Publié: (2025)

Dynamic neuron approach to deep neural networks: Decoupling neurons for renormalization group analysis
par: Lee, Donghee, et autres
Publié: (2024)