Skip to content
conftrace_
2021 JMLR JMLR 2021

On lp-hyperparameter Learning via Bilevel Nonsmooth Optimization

Abstract

We propose a bilevel optimization strategy for selecting the best hyperparameter value for the nonsmooth $\ell_p$ regularizer with $0 [abs][pdf][bib] © JMLR 2021. (edit, beta)

🌉 Interdisciplinary Bridge - Artificial Intelligence and Mathematics & Optimization
📈 Trend Setter - Bilevel Optimization

Authors

Takayuki Okuno, Akiko Takeda, Akihiro Kawana, Motokazu Watanabe

Topics

Machine Learning > Core Methods > Representation Learning Machine Learning > Optimization & Theory > Optimization Mathematics & Optimization > Optimization > Continuous Optimization Machine Learning > Learning Types > Hyperparameter Optimization Mathematics & Optimization > Optimization > Bilevel Optimization

Keywords

hyperparameter learning nonsmooth optimization bilevel optimization ℓp regularizer lp regularizer
Download PDF

More from JMLR 2021

All JMLR 2021 papers →
Wasserstein distance estimates for the distributions of numerical approximations to ergodic stochastic differential equations
Quasi-Monte Carlo Quasi-Newton in Variational Bayes
Consistency of Gaussian Process Regression in Metric Spaces
Mixture Martingales Revisited with Applications to Sequential Tests and Confidence Intervals
Statistical Guarantees for Local Spectral Clustering on Random Neighborhood Graphs
Statistically and Computationally Efficient Change Point Localization in Regression Settings