Learning dynamic systems from time-series data an application to gene regulatory networks
Change log
Authors
Timoteo, IJPM
Holden, SB
Abstract
We propose a local search approach for learning dynamic systems from time-series data, using networks of differential equations as the underlying model. We evaluate the performance of our approach for two scenarios: first, by comparing with an l1-regularization approach under the assumption of a uniformly weighted network for identifying systems of masses and springs; and then on the task of learning gene regulatory networks, where we compare it with the best performers in the DREAM4 challenge using the original dataset for that challenge. Our method consistently improves on the performance of the other methods considered in both scenarios.
Description
Keywords
Graphical Models, Local Search, Bioinformatics Application
Journal Title
ICPRAM 2015 - 4th International Conference on Pattern Recognition Applications and Methods, Proceedings
Conference Name
International Conference on Pattern Recognition Applications and Methods
Journal ISSN
Volume Title
2
Publisher
SCITEPRESS - Science and and Technology Publications
Publisher DOI
Sponsorship
Ivo Timoteo is supported by an FCT Individual Doctoral Fellowship, number SFRH/BD/88466/2012.