This article investigates the use of extended Kalman filtering to train recurrent neural networks with rather general convex loss functions and regularization terms on the network parameters, including ℓ1-regularization. We show that the learning method is competitive with respect to stochastic gradient descent in a nonlinear system identification benchmark and in training a linear system with binary outputs. We also explore the use of the algorithm in data-driven nonlinear model predictive control and its relation with disturbance models for offset-free closed-loop tracking. © 1963-2012 IEEE.

Recurrent Neural Network Training with Convex Loss and Regularization Functions by Extended Kalman Filtering

Bemporad, A.
2023-01-01

Abstract

This article investigates the use of extended Kalman filtering to train recurrent neural networks with rather general convex loss functions and regularization terms on the network parameters, including ℓ1-regularization. We show that the learning method is competitive with respect to stochastic gradient descent in a nonlinear system identification benchmark and in training a linear system with binary outputs. We also explore the use of the algorithm in data-driven nonlinear model predictive control and its relation with disturbance models for offset-free closed-loop tracking. © 1963-2012 IEEE.
2023
Bandpass filters
Extended Kalman filters
Functions
Linear systems
Model predictive control
Nonlinear systems
Random processes
Recurrent neural networks
File in questo prodotto:
File Dimensione Formato  
Recurrent_Neural_Network_Training_With_Convex_Loss_and_Regularization_Functions_by_Extended_Kalman_Filtering.pdf

non disponibili

Tipologia: Versione Editoriale (PDF)
Licenza: Copyright dell'editore
Dimensione 656.97 kB
Formato Adobe PDF
656.97 kB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.11771/27818
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
social impact