ICML 2019oral28 citations
Efficient optimization of loops and limits with randomized telescoping sums
Abstract
We consider optimization problems in which the objective requires an inner loop with many steps or is the limit of a sequence of increasingly costly approximations. Meta-learning, training recurrent neural networks, and optimization of the solutions to differential equations are all examples of optimization problems with this character. In such problems, it can be expensive to compute the objective function value and its gradient, but truncating the loop or using less accurate approximations can induce biases that damage the overall solution. We propose
BibTeX
@InProceedings{pmlr-v97-beatson19a,
title = {Efficient optimization of loops and limits with randomized telescoping sums},
author = {Beatson, Alex and Adams, Ryan P},
booktitle = {Proceedings of the 36th International Conference on Machine Learning},
pages = {534--543},
year = {2019},
editor = {Chaudhuri, Kamalika and Salakhutdinov, Ruslan},
volume = {97},
series = {Proceedings of Machine Learning Research},
month = {09--15 Jun},
publisher = {PMLR},
pdf = {http://proceedings.mlr.press/v97/beatson19a/beatson19a.pdf},
url = {https://proceedings.mlr.press/v97/beatson19a.html},
abstract = {We consider optimization problems in which the objective requires an inner loop with many steps or is the limit of a sequence of increasingly costly approximations. Meta-learning, training recurrent neural networks, and optimization of the solutions to differential equations are all examples of optimization problems with this character. In such problems, it can be expensive to compute the objective function value and its gradient, but truncating the loop or using less accurate approximations can induce biases that damage the overall solution. We propose