Importance Sampling for Deep System Identification

Abstract : This paper revisit the methodology of system identification and shows how new paradigms from machine learning can be used to improve the model identification performance in the case of non-linear systems observed with noisy and unbalanced dataset. We prove that using importance sampling schemes in system identification can provide significant performance boost on a wide variety of systems, in particular when some of the system dynamic is only exhibited by relatively rare events. The performance of the approaches is evaluated on a real and simulated drone and two standard datasets from real robotic systems. Our approach consistently outperforms baseline approaches on these datasets, all the more when the datasets are noisy and unbalanced.
Document type :
Preprints, Working Papers, ...
Complete list of metadatas

Cited literature [21 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-02278171
Contributor : Cedric Pradalier <>
Submitted on : Wednesday, September 4, 2019 - 10:41:00 AM
Last modification on : Tuesday, September 17, 2019 - 11:44:06 AM

File

RAL-IROS.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02278171, version 1

Citation

Antoine Mahé, Antoine Richard, Benjamin Mouscadet, Matthieu Geist, Cedric Pradalier. Importance Sampling for Deep System Identification. 2019. ⟨hal-02278171⟩

Share

Metrics

Record views

46

Files downloads

79