Autonomous learning of parameters in differential equations

Adel Mezine; Artémis Llamosi; Veronique Letort; Michele Sebag; Florence d'Alché-Buc

Autonomous learning of parameters in differential equations

Adel Mezine ¹ Artémis Llamosi ² Veronique Letort ³ Michele Sebag ⁴ Florence d'Alché-Buc ^{1, 5}

1 IBISC - Informatique, Biologie Intégrative et Systèmes Complexes

2 UPD7 - Université Paris Diderot - Paris 7

3 MAS - Mathématiques Appliquées aux Systèmes - EA 4037

4 LRI - Laboratoire de Recherche en Informatique

5 LTCI - Laboratoire Traitement et Communication de l'Information

Abstract : We propose EDEN+ 1 a fully automatic learner of parameters in dynamical systems that selects automatically the next experiment to do in the laboratory to improve its performance. EDEN+ improves upon EDEN, an experimental design algorithm proposed in the context of DREAM 6 and 7 challenges, with several new features: ability to take into account experiments with different costs, Monte-Carlo Tree Search parallelization, global optimization for parameter estimation. An illustration of the behaviour of EDEN+ is given on one of the DREAM7 challenging problems.

Keywords : active learning system identification Monte-Carlo Tree Search multi-armed bandits gene regulatory networks differential equations

Document type :

Conference papers

Domain :

Mathematics [math] / Dynamical Systems [math.DS]

Life Sciences [q-bio] / Genetics

Complete list of metadatas

Cited literature [12 references] Display Hide Download

https://hal-ecp.archives-ouvertes.fr/hal-01272370
Contributor : Veronique Letort <>
Submitted on : Wednesday, February 10, 2016 - 4:54:31 PM
Last modification on : Thursday, December 19, 2019 - 1:12:37 AM
Long-term archiving on: Saturday, November 12, 2016 - 5:20:18 PM

AutoML_soumis.pdf

Files produced by the author(s)

Autonomous learning of parameters in differential equations

File

Identifiers

Collections

Citation

Export

Metrics

1037

61

Contact

Informations

Autonomous learning of parameters in differential equations

File

Identifiers

Collections

Citation

Export

Share

Metrics

1037

61

Contact

Informations