ICML 2016poster51 citations
Starting Small - Learning with Adaptive Sample Sizes
Hadi Daneshmand, Aurelien Lucchi, Thomas Hofmann
Abstract
For many machine learning problems, data is abundant and it may be prohibitive to make multiple passes through the full training set. In this context, we investigate strategies for dynamically increasing the effective sample size, when using iterative methods such as stochastic gradient descent. Our interest is motivated by the rise of variance-reduced methods, which achieve linear convergence rates that scale favorably for smaller sample sizes. Exploiting this feature, we show - theoretically and empirically - how to obtain significant speed-ups with a novel algorithm that reaches statistical accuracy on an n-sample in 2n, instead of n log n steps.
BibTeX
@InProceedings{pmlr-v48-daneshmand16,
title = {Starting Small - Learning with Adaptive Sample Sizes},
author = {Daneshmand, Hadi and Lucchi, Aurelien and Hofmann, Thomas},
booktitle = {Proceedings of The 33rd International Conference on Machine Learning},
pages = {1463--1471},
year = {2016},
editor = {Balcan, Maria Florina and Weinberger, Kilian Q.},
volume = {48},
series = {Proceedings of Machine Learning Research},
address = {New York, New York, USA},
month = {20--22 Jun},
publisher = {PMLR},
pdf = {http://proceedings.mlr.press/v48/daneshmand16.pdf},
url = {https://proceedings.mlr.press/v48/daneshmand16.html},
abstract = {For many machine learning problems, data is abundant and it may be prohibitive to make multiple passes through the full training set. In this context, we investigate strategies for dynamically increasing the effective sample size, when using iterative methods such as stochastic gradient descent. Our interest is motivated by the rise of variance-reduced methods, which achieve linear convergence rates that scale favorably for smaller sample sizes. Exploiting this feature, we show - theoretically and empirically - how to obtain significant speed-ups with a novel algorithm that reaches statistical accuracy on an n-sample in 2n, instead of n log n steps.}
}