Human-in-the-loop Active Covariance Learning for Improving Prediction in Small Data Sets

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Standard

Human-in-the-loop Active Covariance Learning for Improving Prediction in Small Data Sets. / Afrabandpey, Homayun; Peltola, Tomi; Kaski, Samuel.

28th International Joint Conference on Artificial Intelligence. 2019.

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Harvard

APA

Afrabandpey, H., Peltola, T., & Kaski, S. (Accepted/In press). Human-in-the-loop Active Covariance Learning for Improving Prediction in Small Data Sets. In 28th International Joint Conference on Artificial Intelligence

Vancouver

Author

Bibtex - Download

@inproceedings{7b3a8fa3c022467691421c216bce6d3a,
title = "Human-in-the-loop Active Covariance Learning for Improving Prediction in Small Data Sets",
abstract = "Learning predictive models from small high-dimensional data sets is a key problem in high-dimensional statistics. Expert knowledge elicitation can help, and a strong line of work focuses on directly eliciting informative prior distributions for parameters. This either requires considerable statistical expertise or is laborious, as the emphasis has been on accuracy and not on efficiency of the process. Another line of work queries about importance of features one at a time, assuming them to be independent and hence missing covariance information. In contrast, we propose eliciting expert knowledge about pairwise feature similarities, to borrow statistical strength in the predictions, and using sequential decision making techniques to minimize the effort of the expert. Empirical results demonstrate improvement in predictive performance on both simulated and real data, in high-dimensional linear regression tasks, where we learn the covariance structure with a Gaussian process, based on sequential elicitation.",
author = "Homayun Afrabandpey and Tomi Peltola and Samuel Kaski",
year = "2019",
month = "5",
language = "English",
booktitle = "28th International Joint Conference on Artificial Intelligence",

}

RIS - Download

TY - GEN

T1 - Human-in-the-loop Active Covariance Learning for Improving Prediction in Small Data Sets

AU - Afrabandpey, Homayun

AU - Peltola, Tomi

AU - Kaski, Samuel

PY - 2019/5

Y1 - 2019/5

N2 - Learning predictive models from small high-dimensional data sets is a key problem in high-dimensional statistics. Expert knowledge elicitation can help, and a strong line of work focuses on directly eliciting informative prior distributions for parameters. This either requires considerable statistical expertise or is laborious, as the emphasis has been on accuracy and not on efficiency of the process. Another line of work queries about importance of features one at a time, assuming them to be independent and hence missing covariance information. In contrast, we propose eliciting expert knowledge about pairwise feature similarities, to borrow statistical strength in the predictions, and using sequential decision making techniques to minimize the effort of the expert. Empirical results demonstrate improvement in predictive performance on both simulated and real data, in high-dimensional linear regression tasks, where we learn the covariance structure with a Gaussian process, based on sequential elicitation.

AB - Learning predictive models from small high-dimensional data sets is a key problem in high-dimensional statistics. Expert knowledge elicitation can help, and a strong line of work focuses on directly eliciting informative prior distributions for parameters. This either requires considerable statistical expertise or is laborious, as the emphasis has been on accuracy and not on efficiency of the process. Another line of work queries about importance of features one at a time, assuming them to be independent and hence missing covariance information. In contrast, we propose eliciting expert knowledge about pairwise feature similarities, to borrow statistical strength in the predictions, and using sequential decision making techniques to minimize the effort of the expert. Empirical results demonstrate improvement in predictive performance on both simulated and real data, in high-dimensional linear regression tasks, where we learn the covariance structure with a Gaussian process, based on sequential elicitation.

UR - https://arxiv.org/abs/1902.09834

M3 - Conference contribution

BT - 28th International Joint Conference on Artificial Intelligence

ER -

ID: 33776797