paper-with-me

홈 › Papers

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

2021-06-07 · Bruno Loureiro, Gabriele Sicuro, Cédric Gerbelot, Alessandro Pacco, Florent Krzakala, Lenka Zdeborová

Generalised linear models for multi-class classification problems are one of the fundamental building blocks of modern machine learning tasks. In this manuscript, we characterise the learning of a mixture of $K$ Gaussians with generic means and covariances via empirical risk minimisation (ERM) with any convex loss and regularisation. In particular, we prove exact asymptotics characterising the ERM estimator in high-dimensions, extending several previous results about Gaussian mixture classification in the literature. We exemplify our result in two tasks of interest in statistical learning: a) classification for a mixture with sparse means, where we study the efficiency of $\ell_1$ penalty with respect to $\ell_2$; b) max-margin multi-class classification, where we characterise the phase transition on the existence of the multi-class logistic maximum likelihood estimator for $K>2$. Finally, we discuss how our theory can be applied beyond the scope of synthetic data, showing that in different cases Gaussian mixtures capture closely the learning curve of classification tasks in real data sets.

📄 PDF Abstract BibTeX arXiv:2106.03791

Code (2)

IdePHICS/GaussMixtureProject 공식 구현
gsicuro/GaussMixtureProject

Tasks

ClassificationMulti-class ClassificationVocal Bursts Intensity Prediction

Similar Papers 제목 키워드 기반

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

2021-12-01 · NeurIPS 2021 12 · Bruno Loureiro, Gabriele Sicuro, Cedric Gerbelot, Alessandro Pacco 외

Generalised linear models for multi-class classification problems are one of the fundamental building blocks of modern machine learning tasks. In this manuscript, we characterise the learning of a mixture of $K$ Gaussian…

ClassificationMulti-class Classification

Asymptotics of Linear Regression with Linearly Dependent Data

2024-12-04 · Behrad Moniri, Hamed Hassani

In this paper we study the asymptotics of linear regression in settings with non-Gaussian covariates where the covariates exhibit a linear dependency structure, departing from the standard assumption of independence. We …

regression

The Breakdown of Gaussian Universality in Classification of High-dimensional Linear Factor Mixtures

2024-10-08 · Xiaoyi Mai, Zhenyu Liao

The assumption of Gaussian or Gaussian mixture data has been extensively exploited in a long series of precise performance analyses of machine learning (ML) methods, on large datasets having comparably numerous samples a…

Risk Bounds for Over-parameterized Maximum Margin Classification on Sub-Gaussian Mixtures

2021-04-28 · NeurIPS 2021 12 · Yuan Cao, Quanquan Gu, Mikhail Belkin

Modern machine learning systems such as deep neural networks are often highly over-parameterized so that they can fit the noisy training data exactly, yet they can still achieve small test errors in practice. In this pap…

ClassificationGeneral Classificationregression

Small-Variance Asymptotics for Exponential Family Dirichlet Process Mixture Models

2012-12-01 · NeurIPS 2012 12 · Ke Jiang, Brian Kulis, Michael. I. Jordan

Links between probabilistic and non-probabilistic learning algorithms can arise by performing small-variance asymptotics, i.e., letting the variance of particular distributions in a graphical model go to zero. For instan…

Clustering