Publications

IdePHICS is a new lab in EPFL, having just been created in September 2020. For full list of publication of Prof. Florent Krzakala, please visit florentkrzakala.com

Warning

Please note that the publication lists from Infoscience integrated into the EPFL website, lab or people pages are frozen following the launch of the new version of platform. The owners of these pages are invited to recreate their publication list from Infoscience. For any assistance, please consult the Infoscience help or contact support.

2024

On the atypical solutions of the symmetric binary perceptron

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

Journal Of Physics A-Mathematical And Theoretical

2024-05-10

Vol. 57 , num. 19, p. 195202.

DOI : 10.1088/1751-8121/ad3a4a

Publications

2024

On the atypical solutions of the symmetric binary perceptron

Gaussian universality of perceptrons with random labels

Statistical mechanics of the maximum-average submatrix problem

2023

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

Multi-layer state evolution under random convolutional design

Theoretical characterization of uncertainty in high-dimensional linear classification

Bayesian reconstruction of memories stored in neural networks from their connectivity

Fluctuations, bias, variance and ensemble of learners: exact asymptotics for convex losses in high-dimension

Artificial Neural Network Training on an Optical Processor via Direct Feedback Alignment

Error scaling laws for kernel classification under source and capacity conditions

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

Expectation consistency for calibration of neural networks

Compressed sensing with ℓ 0-norm: statistical physics analysis & algorithms for signal recovery

On double-descent in uncertainty quantification in overparametrized models

Tree-AMP: Compositional Inference with Tree Approximate Message Passing

2022

Asymptotic Errors for Teacher-Student Convex Generalized Linear Models (Or: How to Prove Kabashima’s Replica Formula)

Generalization error rates in kernel regression: the crossover from the noiseless to noisy regime*

Perturbative construction of mean-field equations in extensive-rank matrix factorization and denoising

Multi-layer State Evolution Under Random Convolutional Design

Subspace clustering in high-dimensions: Phase transitions & Statistical-to-Computational gap

Phase diagram of Stochastic Gradient Descent in high-dimensional two-layer neural networks

Optimal denoising of rotationally invariant rectangular matrices

Fluctuations, Bias, Variance & Ensemble of Learners: Exact Asymptotics for Convex Losses in High-Dimension

Adversarial Robustness by Design Through Analog Computing And Synthetic Gradients

2021

The Gaussian equivalence of generative models for learning with shallow neural networks

Dynamical mean-field theory for stochastic gradient descent in Gaussian mixture classification*

Generalisation error in learning with random features and the hidden manifold model*

Construction of optimal spectral methods in phase retrieval

Epidemic mitigation by statistical inference from contact tracing data

Classifying high-dimensional Gaussian mixtures: Where kernel methods fail and neural networks succeed

Learning Gaussian Mixtures with Generalized Linear Models: Precise Asymptotics in High-dimensions

Generalization Error Rates in Kernel Regression: The Crossover from the Noiseless to Noisy Regime

Learning curves of generic features maps for realistic datasets with a teacher-student model

2020

The Spiked Matrix Model With Generative Priors

Mutual Information and Optimality of Approximate Message-Passing in Random Linear Estimation

Generalization error in high-dimensional perceptrons: Approaching Bayes error with convex optimization

Phase retrieval in high dimensions: Statistical and computational phase transitions

Dynamical mean-field theory for stochastic gradient descent in Gaussian mixture classification

Direct Feedback Alignment Scales to Modern Deep Learning Tasks and Architectures

Complex Dynamics in Simple Neural Networks: Understanding Gradient Flow in Phase Retrieval

Reservoir Computing meets Recurrent Kernels and Structured Transforms

2019

Entropy and mutual information in models of deep neural networks

The committee machine: computational to statistical gaps in learning a two-layers neural network

Optimal errors and phase transitions in high-dimensional generalized linear models

2018

Entropy and mutual information in models of deep neural networks

The committee machine: Computational to statistical gaps in learning a two-layers neural network

The Mutual Information in Random Linear Estimation Beyond i.i.d. Matrices

2016

Scampi: a robust approximate message-passing framework for compressive imaging