We provide a novel transcription of monotone operator theory to the non-Euclidean finite-dimensional spaces ℓ 1 and ℓ ∞ . We first establish properties of mappings which are monotone with respect to the non-Euclidean norms ℓ 1 or ℓ ∞ . In analogy with their Euclidean counterparts, mappings which are monotone with respect to a non-Euclidean norm are amenable to numerous algorithms for computing their zeros. We demonstrate that several classic iterative methods for computing zeros of monotone operators are directly applicable in the non-Euclidean framework. We present a case-study in the equilibrium computation of recurrent neural networks and demonstrate that casting the computation as a suitable operator splitting problem improves convergence rates.

Non-Euclidean Monotone Operator Theory with Applications to Recurrent Neural Networks / Davydov, Alexander; Jafarpour, Saber; Proskurnikov, Anton V.; Bullo, Francesco. - ELETTRONICO. - (2022), pp. 6332-6337. (Intervento presentato al convegno 2022 IEEE 61st Conference on Decision and Control (CDC) tenutosi a Cancun, Mexico nel 06-09 December 2022) [10.1109/CDC51059.2022.9993197].

Non-Euclidean Monotone Operator Theory with Applications to Recurrent Neural Networks

Proskurnikov, Anton V.;Bullo, Francesco
2022

Abstract

We provide a novel transcription of monotone operator theory to the non-Euclidean finite-dimensional spaces ℓ 1 and ℓ ∞ . We first establish properties of mappings which are monotone with respect to the non-Euclidean norms ℓ 1 or ℓ ∞ . In analogy with their Euclidean counterparts, mappings which are monotone with respect to a non-Euclidean norm are amenable to numerous algorithms for computing their zeros. We demonstrate that several classic iterative methods for computing zeros of monotone operators are directly applicable in the non-Euclidean framework. We present a case-study in the equilibrium computation of recurrent neural networks and demonstrate that casting the computation as a suitable operator splitting problem improves convergence rates.
2022
978-1-6654-6761-2
File in questo prodotto:
File Dimensione Formato  
Non-Euclidean_Monotone_Operator_Theory_with_Applications_to_Recurrent_Neural_Networks.pdf

accesso riservato

Tipologia: 2a Post-print versione editoriale / Version of Record
Licenza: Non Pubblico - Accesso privato/ristretto
Dimensione 1.22 MB
Formato Adobe PDF
1.22 MB Adobe PDF   Visualizza/Apri   Richiedi una copia
main.pdf

accesso aperto

Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: Pubblico - Tutti i diritti riservati
Dimensione 511.28 kB
Formato Adobe PDF
511.28 kB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

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/11583/2974559