Framing RNN as a kernel method: A neural ODE approach - Archive ouverte HAL
Communication Dans Un Congrès Année : 2021

Framing RNN as a kernel method: A neural ODE approach

Résumé

Building on the interpretation of a recurrent neural network (RNN) as a continuoustime neural differential equation, we show, under appropriate conditions, that the solution of a RNN can be viewed as a linear function of a specific feature set of the input sequence, known as the signature. This connection allows us to frame a RNN as a kernel method in a suitable reproducing kernel Hilbert space. As a consequence, we obtain theoretical guarantees on generalization and stability for a large class of recurrent networks. Our results are illustrated on simulated datasets.
Fichier principal
Vignette du fichier
2106.01202.pdf (821.31 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03943120 , version 1 (10-02-2023)

Identifiants

Citer

Adeline Fermanian, Pierre Marion, Jean-Philippe Vert, Gérard Biau. Framing RNN as a kernel method: A neural ODE approach. Thirty-fifth Conference on Neural Information Processing Systems, Dec 2021, Virtual-only, United States. ⟨10.48550/arXiv.2106.01202⟩. ⟨hal-03943120⟩
39 Consultations
56 Téléchargements

Altmetric

Partager

More