About: Scalable Identification of Partially Observed Systems with Certainty-Equivalent EM

Facets (new session)
Description
Metadata
Settings
- owl:sameAs
- Inference Rule:

About: Scalable Identification of Partially Observed Systems with Certainty-Equivalent EM Goto Sponge NotDistinct Permalink

An Entity of Type : schema:ScholarlyArticle, within Data Space : wasabi.inria.fr associated with source document(s)

Attributes	Values
type	Academic Article research paper schema:ScholarlyArticle
isDefinedBy	Covid-on-the-Web dataset
title	Scalable Identification of Partially Observed Systems with Certainty-Equivalent EM
Creator	De Becdelièvre, Jean Gupta, Jayesh Kochenderfer, Mykel Kroo, Ilan Manchester, Zachary Menda, Kunal
source	ArXiv
abstract	System identification is a key step for model-based control, estimator design, and output prediction. This work considers the offline identification of partially observed nonlinear systems. We empirically show that the certainty-equivalent approximation to expectation-maximization can be a reliable and scalable approach for high-dimensional deterministic systems, which are common in robotics. We formulate certainty-equivalent expectation-maximization as block coordinate-ascent, and provide an efficient implementation. The algorithm is tested on a simulated system of coupled Lorenz attractors, demonstrating its ability to identify high-dimensional systems that can be intractable for particle-based approaches. Our approach is also used to identify the dynamics of an aerobatic helicopter. By augmenting the state with unobserved fluid states, a model is learned that predicts the acceleration of the helicopter better than state-of-the-art approaches. The codebase for this work is available at https://github.com/sisl/CEEM.
has issue date	2020-06-20(xsd:dateTime)
has license	arxiv
sha1sum (hex)	6afa374dbda286f5e43d21e80252ddc62cb7e7ab
resource representing a document's title	Scalable Identification of Partially Observed Systems with Certainty-Equivalent EM
resource representing a document's body	covid:6afa374dbda286f5e43d21e80252ddc62cb7e7ab#body_text
is schema:about of	named entity 'approximation' named entity 'partially' named entity 'codebase' named entity 'ACCELERATION' named entity 'LEARNED' named entity 'COUPLED' named entity 'OBSERVED' named entity 'SYSTEM' named entity 'NONLINEAR SYSTEMS' named entity 'ESTIMATOR' named entity 'OUR APPROACH' named entity 'IS A' named entity 'ROBOTICS' named entity 'CERTAINTY' named entity 'COM' named entity 'PREDICTION' named entity 'This' named entity 'control' named entity 'expectation-maximization' named entity 'nonlinear systems' named entity 'estimator' named entity 'deterministic systems' named entity 'expectation-maximization' named entity 'training dataset' named entity 'H25' named entity 'radial basis functions' named entity 'vector' named entity 'Hessian matrix' named entity 'Runge-Kutta method' named entity 'deterministic' named entity 'LSTM' named entity 'inference' named entity 'chaotic behavior' named entity 'Gaussian' named entity 'identification problem' named entity 'nonlinear systems' named entity 'EM algorithm' named entity 'Stochastic Approximation' named entity 'EKF' named entity 'RNNs' named entity 'state variables' named entity 'raw data' named entity 'Gaussian' named entity 'RMSE' named entity 'unimodal' named entity 'neural network' named entity 'neural network' named entity 'gray-box' named entity 'iid' named entity 'Subspace Identification' named entity 'RMSE' named entity 'variance' named entity 'numerical stability' named entity 'predictive models' named entity 'EKS' named entity 'gyroscope' named entity 'analytic solution' named entity 'neural network' named entity 'SSM' named entity 'parameter estimates' named entity 'actuators' named entity 'deep neural network' named entity 'deterministic systems' named entity 'randomly distributed' named entity 'non-linear programming' named entity 'non-linear' named entity 'block diagonal' named entity 'velocity' named entity 'non-linear' named entity 'parameter estimation' named entity 'Autoregressive' named entity 'Algorithm' named entity 'magnetometer' named entity 'reference frame' named entity 'normal distribution' named entity 'MATLAB' named entity 'non-linear systems' named entity 'Expectation-Maximization' named entity 'trust-region' named entity 'neural networks' named entity 'Neural networks' named entity 'robotics'

Faceted Search & Find service v1.13.91 as of Mar 24 2020

Alternative Linked Data Documents: Sponger | ODE Content Formats:

RDF

ODATA

Microdata

About

OpenLink Virtuoso version 07.20.3229 as of Jul 10 2020, on Linux (x86_64-pc-linux-gnu), Single-Server Edition (94 GB total memory)
Data on this page belongs to its respective rights holders.
Virtuoso Faceted Browser Copyright © 2009-2025 OpenLink Software