Distance Based Modeling of Interactions in Structured Regression
Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence. Pages 2032-2037.
Abstract
Graphical models, as applied to multi-target prediction problems, commonly utilize interaction terms to impose structure among the output variables. Often, such structure is based on the assumption that related outputs need to be similar and interaction terms that force them to be closer are adopted. Here we relax that assumption and propose a feature that is based on distance and can adapt to ensure that variables have smaller or larger difference in values. We utilized a Gaussian Conditional Random Field model, where we have extended its originally proposed interaction potential to include a distance term. The extended model is compared to the baseline in various structured regression setups. An increase in predictive accuracy was observed on both synthetic examples and real-world applications, including challenging tasks from climate and healthcare domains.
@inproceedings{ijcai2016p290,
title = {Distance Based Modeling of Interactions in Structured Regression},
author = {Stojkovic, Ivan and Jelisavcic, Vladisav and Milutinovic, Veljko and Obradovic, Zoran},
booktitle = {Proceedings of the Twenty-Fifth International Joint Conference on
Artificial Intelligence, {IJCAI-16}},
publisher = {AAAI Press / International Joint Conferences on Artificial Intelligence},
editor = {Subbarao Kambhampati},
pages = {2032--2037},
year = {2016},
month = {7},
url = {https://ijcai.org/proceedings/2016/290},
}
TY - CPAPER TI - Distance Based Modeling of Interactions in Structured Regression AU - Stojkovic, Ivan AU - Jelisavcic, Vladisav AU - Milutinovic, Veljko AU - Obradovic, Zoran T2 - Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-16 PB - AAAI Press / International Joint Conferences on Artificial Intelligence PY - 2016 SP - 2032 EP - 2037 UR - https://ijcai.org/proceedings/2016/290 ER -Copied
Machine-readable versions of this paper, for tools and AI agents: