Statistical Physics Analysis of Graph Neural Networks: Approaching Optimality in the Contextual Stochastic Block Model

O. Duranthon; L. Zdeborova 

Optimal Thresholds and Algorithms for a Model of Multi-modal Learning in High Dimensions

C. Keup; L. Zdeborova 

Dynamical cavity method for hypergraphs and its application to quenches in the k-XOR-SAT problem

A. Maier; F. Behrens; L. Zdeborová 

A phase transition between positional and semantic learning in a solvable model of dot-product attention

H. Cui; F. Behrens; F. Krzakala; L. Zdeborová 

Bilinear Sequence Regression: A Model for Learning from Long Sequences of High-Dimensional Tokens

V. Erba; E. Troiani; L. Biggio; A. Maillard; L. Zdeborová 

Low-rank matrix estimation with inhomogeneous noise

A. Guionnet; J. Ko; F. Krzakala; L. Zdeborová 

Phase Diagram of Compressed Sensing with l₀-norm Regularization

D. Barbier; C. Lucibello; L. Saglietti; F. Krzakala; L. Zdeborova 

Bayes-optimal learning of deep random networks of extensive-width*

H. Cui; F. Krzakala; L. Zdeborova 

Integer traffic assignment problem: Algorithms and insights on random graphs

R. Harfouche; G. Piccioli; L. Zdeborová 

Estimating Rank-One Matrices with Mismatched Prior and Noise: Universality and Large Deviations

A. Guionnet; J. Ko; F. Krzakala; L. Zdeborová 

Building Conformal Prediction Intervals with Approximate Message Passing

L. A. Clarte; L. Zdeborová 

Counting in Small Transformers: The Delicate Interplay between Attention and Feed-Forward Layers

F. Behrens; L. Zdeborová; L. Biggio 

Fundamental limits of learning in sequence multi-index models and deep attention networks: high-dimensional asymptotics and sharp thresholds

E. Troiani; H. C. Cui; Y. Dandi; F. Krzakala; L. Zdeborová 

Fundamental computational limits of weak learnability in high-dimensional multi-index models

E. Troiani; Y. Dandi; L. Defilippis; L. Zdeborová; B. Loureiro et al. 

Structure and Computation in Disordered Systems and Neural Networks

Theoretical characterization of uncertainty in high-dimensional machine learning

Counting and hardness-of-finding fixed points in cellular automata on random graphs

C. X. Koller; F. Behrens; L. Zdeborová 

Les Houches 2022 Special Issue: Editorial

F. Krzakala; L. Zdeborova 

Quenches in the Sherrington-Kirkpatrick model

V. Erba; F. Behrens; F. Krzakala; L. Zdeborová 

Sampling with flows, diffusion, and autoregressive neural networks from a spin-glass perspective

D. Ghio; Y. Dandi; F. Krzakala; L. Zdeborová 

On the atypical solutions of the symmetric binary perceptron

D. Barbier; A. El Alaoui; F. Krzakala; L. Zdeborova 

Rigorous Dynamical Mean-Field Theory for Stochastic Gradient Descent Methods

C. Gerbelot; E. Troiani; F. Mignacco; F. Krzakala; L. Zdeborová 

Dynamical phase transitions in graph cellular automata

F. Behrens; B. Hudcová; L. Zdeborová 

Optimal Inference in Contextual Stochastic Block Models

O. Duranthon; L. Zdeborová 

Gibbs sampling the posterior of neural networks

G. Piccioli; E. Troiani; L. Zdeborova 

Gaussian universality of perceptrons with random labels

F. Gerace; F. Krzakala; B. Loureiro; L. Stephan; L. Zdeborova 

Statistical mechanics of the maximum-average submatrix problem

V. Erba; F. Krzakala; R. Perez Ortiz; L. Zdeborova 

Universality laws for Gaussian mixtures in generalized linear models

Y. Dandi; L. Stephan; F. Krzakala; B. Loureiro; L. Zdeborová 

Asymptotic generalization error of a single-layer graph convolutional network

O. Duranthon; L. Zdeborová 

Analysis of Bootstrap and Subsampling in High-dimensional Regularized Regression

L. Clarte; A. Vandenbroucque; G. Dalle; B. Loureiro; F. Krzakala et al. 

ANALYSIS OF LEARNING A FLOW-BASED GENERATIVE MODEL FROM LIMITED SAMPLE COMPLEXITY

H. C. Cui; F. Krzakala; E. Vanden-Eijnden; L. Zdeborová 

Bayes-optimal learning of an extensive-width neural network from quadratically many samples

A. Maillard; E. Troiani; S. Martin; L. Zdeborová; F. Krzakala 

Asymptotics of feature learning in two-layer networks after one gradient-step

H. C. Cui; L. Pesce; Y. Dandi; F. Krzakala; Y. Lu et al. 

Advances in Algorithms for Sampling, Optimization and Inference in Disordered Systems

Topics in statistical physics of high-dimensional machine learning

The Benefits of Reusing Batches for Gradient Descent in Two-Layer Networks: Breaking the Curse of Information and Leap Exponents

Y. Dandi; E. Troiani; L. Arnaboldi; L. Pesce; L. Zdeborová et al. 

Phase diagram of stochastic gradient descent in high-dimensional two-layer neural networks

R. Veiga; L. Stephan; B. Loureiro; F. Krzakala; L. Zdeborova 

Multi-layer state evolution under random convolutional design

M. Daniels; C. Gerbelot; F. Krzakala; L. Zdeborova 

Bayes-optimal inference for spreading processes on random networks

D. Ghio; A. L. M. Aragon; I. Biazzo; L. Zdeborova 

Neural-prior stochastic block model

O. Duranthon; L. Zdeborová 

Backtracking Dynamical Cavity Method

F. Behrens; B. Hudcová; L. Zdeborová 

Theoretical characterization of uncertainty in high-dimensional linear classification

L. Clarte; B. Loureiro; F. Krzakala; L. Zdeborova 

Learning curves for the multi-class teacher-student perceptron

E. Cornacchia; F. Mignacco; R. Veiga; C. Gerbelot; B. Loureiro et al. 

Bayesian reconstruction of memories stored in neural networks from their connectivity

S. Goldt; F. Krzakala; L. Zdeborová; N. Brunel 

Tree-AMP: Compositional Inference with Tree Approximate Message Passing

A. Baker; B. Aubin; F. Krzakala; L. Zdeborová 

Error scaling laws for kernel classification under source and capacity conditions

H. C. Cui; B. Loureiro; F. Krzakala; L. Zdeborová 

High-dimensional Asymptotics of Denoising Autoencoders

H. C. Cui; L. Zdeborová 

Bayes-optimal Learning of Deep Random Networks of Extensive-width

H. C. Cui; F. Krzakala; L. Zdeborová 

On double-descent in uncertainty quantification in overparametrized models

L. A. Clarte; B. Loureiro; F. Krzakala; L. Zdeborová 

Expectation consistency for calibration of neural networks

L. A. Clarte; B. Loureiro; F. Krzakala; L. Zdeborová