Adaptive Subgradient Methods for Online Learning and Stochastic Optimization

Abstract

We present a new family of subgradient methods that dynamically incorporate knowledge of the geometry of the data observed in earlier iterations to perform more informative gradientbased learning. Metaphorically, the adaptation allows us to find needles in haystacks in the form of very predictive but rarely seen features. Our paradigm stems from recent advances in stochastic optimization and online learning which employ proximal functions to control the gradient steps of the algorithm. We describe and analyze an apparatus for adaptively modifying the proximal function, which significantly simplifies setting a learning rate and results in regret guarantees that are provably as good as the best proximal function that can be chosen in hindsight. We give several efficient algorithms for empirical risk minimization problems with common and important regularization functions and domain constraints. We experimentally study our theoretical analysis and show that adaptive subgradient methods outperform state-of-the-art, yet non-adaptive, subgradient algorithms.

View Slides

Extracted Key Phrases

11 Figures and Tables

0500200920102011201220132014201520162017
Citations per Year

2,410 Citations

Semantic Scholar estimates that this publication has 2,410 citations based on the available data.

See our FAQ for additional information.

Cite this paper

@article{Duchi2010AdaptiveSM, title={Adaptive Subgradient Methods for Online Learning and Stochastic Optimization}, author={John C. Duchi and Elad Hazan and Yoram Singer}, journal={Journal of Machine Learning Research}, year={2010}, volume={12}, pages={2121-2159} }