ML Anthology

Experimental Design in Dynamical System Identification: A Bandit-Based Active Learning Approach

Artémis Llamosi, Adel Mezine, Florence d'Alché-Buc, Véronique Letort, Michèle Sebag
ECML-PKDD 2014 pp. 306-321
doi:10.1007/978-3-662-44851-9_20 /ecmlpkdd/2014/llamosi2014ecmlpkdd-experimental/

Abstract

This study focuses on dynamical system identification, with the reverse modeling of a gene regulatory network as motivating application. An active learning approach is used to iteratively select the most informative experiments needed to improve the parameters and hidden variables estimates in a dynamical model given a budget for experiments. The design of experiments under these budgeted resources is formalized in terms of sequential optimization. A local optimization criterion (reward) is designed to assess each experiment in the sequence, and the global optimization of the sequence is tackled in a game-inspired setting, within the Upper Confidence Tree framework combining Monte-Carlo tree-search and multi-armed bandits. The approach, called EDEN for Experimental Design for parameter Estimation in a Network, shows very good performances on several realistic simulated problems of gene regulatory network reverse-modeling, inspired from the international challenge DREAM7.

PDF ECML-PKDD Semantic Scholar

Cite

Text

Llamosi et al. "Experimental Design in Dynamical System Identification: A Bandit-Based Active Learning Approach." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2014. doi:10.1007/978-3-662-44851-9_20

Markdown

[Llamosi et al. "Experimental Design in Dynamical System Identification: A Bandit-Based Active Learning Approach." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2014.](https://mlanthology.org/ecmlpkdd/2014/llamosi2014ecmlpkdd-experimental/) doi:10.1007/978-3-662-44851-9_20

BibTeX

@inproceedings{llamosi2014ecmlpkdd-experimental,
  title     = {{Experimental Design in Dynamical System Identification: A Bandit-Based Active Learning Approach}},
  author    = {Llamosi, Artémis and Mezine, Adel and d'Alché-Buc, Florence and Letort, Véronique and Sebag, Michèle},
  booktitle = {European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases},
  year      = {2014},
  pages     = {306-321},
  doi       = {10.1007/978-3-662-44851-9_20},
  url       = {https://mlanthology.org/ecmlpkdd/2014/llamosi2014ecmlpkdd-experimental/}
}