Scalable Non-linear Learning with Adaptive Polynomial Expansions
Alekh Agarwal ⋅ Alina Beygelzimer ⋅ Daniel Hsu ⋅ John Langford ⋅ Matus J Telgarsky
2014 Poster
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.
Chat is not available.
Successful Page Load