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

Abstract

There has been a recent surge of interest in the study of asymptotic reconstruction performance in various cases of generalized linear estimation problems in the teacher-student setting, especially for the case of i.i.d standard normal matrices. Here, we go beyond these matrices, and prove an analytical formula for the reconstruction performance of convex generalized linear models with rotationally-invariant data matrices with arbitrary bounded spectrum, rigorously confirming, under suitable assumptions, a conjecture originally derived using the replica method from statistical physics. The proof is achieved by leveraging on message passing algorithms and the statistical properties of their iterates, allowing to characterize the asymptotic empirical distribution of the estimator. For sufficiently strongly convex problems, we show that the two-layer vector approximate message passing algorithm (2-MLVAMP) converges, where the convergence analysis is done by checking the stability of an equivalent dynamical system, which gives the result for such problems. We then show that, under a concentration assumption, an analytical continuation may be carried out to extend the result to convex (non-strongly) problems. We illustrate our claim with numerical examples on mainstream learning methods such as sparse logistic regression and linear support vector classifiers, showing excellent agreement between moderate size simulation and the asymptotic prediction.

53 pages, 4 figures

References in corpus (3)

• Generalisation error in learning with random features and the hidden manifold model
• Inference from correlated patterns: a unified theory for perceptron learning and linear vector channels
• Macroscopic Analysis of Vector Approximate Message Passing in a Model Mismatch Setting

Cited by in corpus (9)

• PCA Initialization for Approximate Message Passing in Rotationally Invariant Models
• Macroscopic Analysis of Vector Approximate Message Passing in a Model Mismatch Setting
• Compressed sensing radar detectors under the row-orthogonal design model: a statistical mechanics perspective
• Strong replica symmetry for high-dimensional disordered log-concave Gibbs measures
• Provable More Data Hurt in High Dimensional Least Squares Estimator
• Learning Gaussian Mixtures with Generalised Linear Models: Precise Asymptotics in High-dimensions
• Optimal generalisation and learning transition in extensive-width shallow neural networks near interpolation
• Statistical mechanics of extensive-width Bayesian neural networks near interpolation
• Statistical physics of deep learning: Optimal learning of a multi-layer perceptron near interpolation