Second-order non-stationary online learning for regression

MaRDI QIDQ5744809zbMATH OpenFDO

Authors Edward Moroshko, Nina Vaits, Koby Crammer

Publication date 19 February 2016

Full work available at URL https://arxiv.org/abs/1303.0140, http://jmlr.csail.mit.edu/papers/v16/moroshko15a.html

zbMATH Keywords

online learningregret boundsnon-stationary input

Mathematics Subject Classification ID

Linear regression; mixed models (62J05) Learning and adaptive systems in artificial intelligence (68T05) Inference from stochastic processes and prediction (62M20)

Recommendations

• Re-adapting the regularization of weights for non-stationary regression
• Adaptive and self-confident on-line learning algorithms
• Online regression with varying Gaussians and non-identical distributions
• A generalized online mirror descent with applications to classification and regression
• Weighted last-step min-max algorithm with improved sub-logarithmic regret

Cited in

(9)

• Nonstationary online convex optimization with multiple predictions
• Non-stationary stochastic optimization
• One step back, two steps forward: interference and learning in recurrent neural networks
• An upper bound for aggregating algorithm for regression with changing dependencies
• Online renewable smooth quantile regression
• Second-Order Online Nonconvex Optimization
• Re-adapting the regularization of weights for non-stationary regression
• Renewable composite quantile method and algorithm for nonparametric models with streaming data
• Recursive ridge regression using second-order stochastic algorithms

Describes a project that uses

Uses Software

• AdaGrad

This page was built for publication: Second-order non-stationary online learning for regression

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5744809)