Image de couverture de livre

A solvable high-dimensional model where nonlinear autoencoders learn structure invisible to PCA while test loss misaligns with generalization

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Mendes, Vicente Conde, Bardone, Lorenzo, Koller, Cédric, Moreira, Jorge Medina, Erba, Vittorio, Troiani, Emanuele, Zdeborová, Lenka
Format:	Preprint
Publié:	2026
Sujets:	Machine Learning Disordered Systems and Neural Networks
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Bayes optimal learning of attention-indexed models
par: Boncoraglio, Fabrizio, et autres
Publié: (2025)

The Nuclear Route: Sharp Asymptotics of ERM in Overparameterized Quadratic Networks
par: Erba, Vittorio, et autres
Publié: (2025)

Bilinear Sequence Regression: A Model for Learning from Long Sequences of High-dimensional Tokens
par: Erba, Vittorio, et autres
Publié: (2024)

Single-Head Attention in High Dimensions: A Theory of Generalization, Weights Spectra, and Scaling Laws
par: Boncoraglio, Fabrizio, et autres
Publié: (2025)

Gibbs Sampling the Posterior of Neural Networks
par: Piccioli, Giovanni, et autres
Publié: (2023)

Counting and Hardness-of-Finding Fixed Points in Cellular Automata on Random Graphs
par: Koller, Cédric, et autres
Publié: (2024)

Rigorous dynamical mean field theory for stochastic gradient descent methods
par: Gerbelot, Cedric, et autres
Publié: (2022)

Scaling Laws and Spectra of Shallow Neural Networks in the Feature Learning Regime
par: Defilippis, Leonardo, et autres
Publié: (2025)

Quenches in the Sherrington-Kirkpatrick model
par: Erba, Vittorio, et autres
Publié: (2024)

Statistical mechanics of the maximum-average submatrix problem
par: Erba, Vittorio, et autres
Publié: (2023)

Minority Takeover in Majority Dynamics: Searching for Rare Initializations via the History Passing Algorithm
par: Jankola, Marek, et autres
Publié: (2025)

Asymptotic Characterisation of Robust Empirical Risk Minimisation Performance in the Presence of Outliers
par: Vilucchio, Matteo, et autres
Publié: (2023)

Bayes-optimal learning of an extensive-width neural network from quadratically many samples
par: Maillard, Antoine, et autres
Publié: (2024)

Fundamental limits of learning in sequence multi-index models and deep attention networks: High-dimensional asymptotics and sharp thresholds
par: Troiani, Emanuele, et autres
Publié: (2025)

The maximum-average subtensor problem: equilibrium and out-of-equilibrium properties
par: Erba, Vittorio, et autres
Publié: (2025)

Fundamental computational limits of weak learnability in high-dimensional multi-index models
par: Troiani, Emanuele, et autres
Publié: (2024)

The Benefits of Reusing Batches for Gradient Descent in Two-Layer Networks: Breaking the Curse of Information and Leap Exponents
par: Dandi, Yatin, et autres
Publié: (2024)

A phase transition between positional and semantic learning in a solvable model of dot-product attention
par: Cui, Hugo, et autres
Publié: (2024)

Multi-layer State Evolution Under Random Convolutional Design
par: Daniels, Mara, et autres
Publié: (2022)

Sliding down the stairs: how correlated latent variables accelerate learning with neural networks
par: Bardone, Lorenzo, et autres
Publié: (2024)

High-dimensional Asymptotics of Denoising Autoencoders
par: Cui, Hugo, et autres
Publié: (2023)

Spectral Thresholds in Correlated Spiked Models and Fundamental Limits of Partial Least Squares
par: Mergny, Pierre, et autres
Publié: (2025)

Building Conformal Prediction Intervals with Approximate Message Passing
par: Clarté, Lucas, et autres
Publié: (2024)

Dataset distillation for memorized data: Soft labels can leak held-out teacher knowledge
par: Behrens, Freya, et autres
Publié: (2025)

Optimal thresholds and algorithms for a model of multi-modal learning in high dimensions
par: Keup, Christian, et autres
Publié: (2024)

Learning from higher-order statistics, efficiently: hypothesis tests, random features, and neural networks
par: Székely, Eszter, et autres
Publié: (2023)

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

Dynamical Cavity Method for Hypergraphs and its Application to Quenches in the k-XOR-SAT Problem
par: Maier, Aude, et autres
Publié: (2024)

Learning with Restricted Boltzmann Machines: Asymptotics of AMP and GD in High Dimensions
par: Xu, Yizhou, et autres
Publié: (2025)

Integer Traffic Assignment Problem: Algorithms and Insights on Random Graphs
par: Harfouche, Rayan, et autres
Publié: (2024)

Counting in Small Transformers: The Delicate Interplay between Attention and Feed-Forward Layers
par: Behrens, Freya, et autres
Publié: (2024)

Maximally-stable Local Optima in Random Graphs and Spin Glasses: Phase Transitions and Universality
par: Dandi, Yatin, et autres
Publié: (2023)

Dynamical Phase Transitions in Graph Cellular Automata
par: Behrens, Freya, et autres
Publié: (2023)

Inference in Spreading Processes with Neural-Network Priors
par: Ghio, Davide, et autres
Publié: (2025)

The Rules-and-Facts Model for Simultaneous Generalization and Memorization in Neural Networks
par: Farné, Gabriele, et autres
Publié: (2026)

On the existence of consistent adversarial attacks in high-dimensional linear classification
par: Vilucchio, Matteo, et autres
Publié: (2025)

Mode mixing and losses in misaligned microcavities
par: Hughes, William J., et autres
Publié: (2023)

Rigorous Asymptotics for First-Order Algorithms Through the Dynamical Cavity Method
par: Dandi, Yatin, et autres
Publié: (2026)

Fundamental limits of Non-Linear Low-Rank Matrix Estimation
par: Mergny, Pierre, et autres
Publié: (2024)

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