conftrace_

2014 NeurIPS NeurIPS 2014

Scalable Non-linear Learning with Adaptive Polynomial Expansions

Abstract

Can we effectively learn a nonlinear representation in time comparable to linear learning? We describe a new algorithm that explicitly and adaptively expands higher-order interaction features over base linear representations. The algorithm is designed for extreme computational efficiency, and an extensive experimental study shows that its computation/prediction tradeoff ability compares very favorably against strong baselines.

📈 Trend Setter - Efficient Computing

🐣 Hot Topic Early Bird - feature engineering

Authors

Alekh Agarwal, Alina Beygelzimer, Daniel J. Hsu, John Langford, Matus J. Telgarsky, Matus J Telgarsky

Topics

Machine Learning > Core Methods > Representation Learning Machine Learning > Optimization & Theory > Optimization Machine Learning > Application Areas > Efficient Computing Machine Learning > Core Methods > Feature Learning Artificial Intelligence > Core AI > Efficient Computing

Keywords

representation learning feature learning adaptive learning efficient computing computational efficiency feature expansion adaptive algorithm nonlinear representation polynomial expansions feature engineering polynomial feature non-linear learning

More from NeurIPS 2014

All NeurIPS 2014 papers →

Searching for Higgs Boson Decay Modes with Deep Learning

A provable SVD-based algorithm for learning topics in dominant admixture corpus

Sparse Random Feature Algorithm as Coordinate Descent in Hilbert Space

New Rules for Domain Independent Lifted MAP Inference

Zeta Hull Pursuits: Learning Nonconvex Data Hulls

Blossom Tree Graphical Models