paper-with-me

홈 › Papers

Landscape Complexity for the Empirical Risk of Generalized Linear Models

2019-12-04 · Antoine Maillard, Gérard Ben Arous, Giulio Biroli

We present a method to obtain the average and the typical value of the number of critical points of the empirical risk landscape for generalized linear estimation problems and variants. This represents a substantial extension of previous applications of the Kac-Rice method since it allows to analyze the critical points of high dimensional non-Gaussian random functions. Under a technical hypothesis, we obtain a rigorous explicit variational formula for the annealed complexity, which is the logarithm of the average number of critical points at fixed value of the empirical risk. This result is simplified, and extended, using the non-rigorous Kac-Rice replicated method from theoretical physics. In this way we find an explicit variational formula for the quenched complexity, which is generally different from its annealed counterpart, and allows to obtain the number of critical points for typical instances up to exponential accuracy.

📄 PDF Abstract BibTeX arXiv:1912.02143

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Landscape Complexity for the Empirical Risk of Generalized Linear Models: Discrimination between Structured Data

2025-03-18 · Theodoros G. Tsironis, Aris L. Moustakas

We use the Kac-Rice formula and results from random matrix theory to obtain the average number of critical points of a family of high-dimensional empirical loss functions, where the data are correlated $d$-dimensional Ga…

The Landscape of Deep Learning Algorithms

2017-05-19 · Pan Zhou, Jiashi Feng

This paper studies the landscape of empirical risk of deep neural networks by theoretically analyzing its convergence behavior to the population risk as well as its stationary points and properties. For an $l$-layer line…

Deep LearningGeneralization Bounds

Localization, Convexity, and Star Aggregation

2021-05-19 · NeurIPS 2021 12 · Suhas Vijaykumar

Offset Rademacher complexities have been shown to provide tight upper bounds for the square loss in a broad class of problems including improper statistical learning and online learning. We show that the offset complexit…

Average Stability is Invariant to Data Preconditioning. Implications to Exp-concave Empirical Risk Minimization

2016-01-15 · Alon Gonen, Shai Shalev-Shwartz

We show that the average stability notion introduced by \cite{kearns1999algorithmic, bousquet2002stability} is invariant to data preconditioning, for a wide class of generalized linear models that includes most of the kn…

Empirical Risk Landscape Analysis for Understanding Deep Neural Networks

2018-01-01 · ICLR 2018 1 · Pan Zhou, Jiashi Feng

This work aims to provide comprehensive landscape analysis of empirical risk in deep neural networks (DNNs), including the convergence behavior of its gradient, its stationary points and the empirical risk itself to the…

Generalization Bounds