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

B Loureiro, C Gerbelot, H Cui, S Goldt… - Advances in …, 2021 - proceedings.neurips.cc
Teacher-student models provide a framework in which the typical-case performance of high-
dimensional supervised learning can be described in closed form. The assumptions of …

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

F Gerace, B Loureiro, F Krzakala… - International …, 2020 - proceedings.mlr.press
We study generalised linear regression and classification for a synthetically generated
dataset encompassing different problems of interest, such as learning with random features …

[BOOK][B] Statistical mechanics of learning

A Engel - 2001 - books.google.com
Learning is one of the things that humans do naturally, and it has always been a challenge
for us to understand the process. Nowadays this challenge has another dimension as we try …

Double trouble in double descent: Bias and variance (s) in the lazy regime

S d'Ascoli, M Refinetti, G Biroli… - … on Machine Learning, 2020 - proceedings.mlr.press
Deep neural networks can achieve remarkable generalization performances while
interpolating the training data. Rather than the U-curve emblematic of the bias-variance …

A statistical-mechanics approach to large-system analysis of CDMA multiuser detectors

T Tanaka - IEEE Transactions on Information theory, 2002 - ieeexplore.ieee.org
We present a theory, based on statistical mechanics, to evaluate analytically the
performance of uncoded, fully synchronous, randomly spread code-division multiple-access …

Are Gaussian data all you need? The extents and limits of universality in high-dimensional generalized linear estimation

L Pesce, F Krzakala, B Loureiro… - … on Machine Learning, 2023 - proceedings.mlr.press
In this manuscript we consider the problem of generalized linear estimation on Gaussian
mixture data with labels given by a single-index model. Our first result is a sharp asymptotic …

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

H Cui, B Loureiro, F Krzakala… - Advances in Neural …, 2021 - proceedings.neurips.cc
In this manuscript we consider Kernel Ridge Regression (KRR) under the Gaussian design.
Exponents for the decay of the excess generalization error of KRR have been reported in …

Triple descent and the two kinds of overfitting: Where & why do they appear?

S d'Ascoli, L Sagun, G Biroli - Advances in neural …, 2020 - proceedings.neurips.cc
A recent line of research has highlighted the existence of a``double descent''phenomenon in
deep learning, whereby increasing the number of training examples N causes the …

A theory of non-linear feature learning with one gradient step in two-layer neural networks

B Moniri, D Lee, H Hassani, E Dobriban - arxiv preprint arxiv:2310.07891, 2023 - arxiv.org
Feature learning is thought to be one of the fundamental reasons for the success of deep
neural networks. It is rigorously known that in two-layer fully-connected neural networks …

Estimating dataset size requirements for classifying DNA microarray data

S Mukherjee, P Tamayo, S Rogers, R Rifkin… - Journal of …, 2003 - liebertpub.com
A statistical methodology for estimating dataset size requirements for classifying microarray
data using learning curves is introduced. The goal is to use existing classification results to …