Skip to Main content Skip to Navigation
Preprints, Working Papers, ...

Continuous Methods : Hamiltonian Domain Translation

Emmanuel Menier 1, 2, 3 Michele Alessandro Bucci 2 Mouadh Yagoubi 3 Lionel Mathelin 1, 4 Marc Schoenauer 2, 1, 5 
4 DATAFLOT - DAtascience, trAnsition, Fluid instability, contrOl, Turbulence
LISN - Laboratoire Interdisciplinaire des Sciences du Numérique, M.-E. - Mécanique-Energétique
5 A&O - A&O (Apprentissage et Optimisation)
LISN - Laboratoire Interdisciplinaire des Sciences du Numérique, SDD - Science des Données, AAC - Algorithmes, Apprentissage et Calcul
Abstract : This paper proposes a novel approach to domain translation. Leveraging established parallels between generative models and dynamical systems, we propose a reformulation of the Cycle-GAN architecture. By embedding our model with a Hamiltonian structure, we obtain a continuous, expressive and most importantly invertible generative model for domain translation.
Document type :
Preprints, Working Papers, ...
Complete list of metadata

https://hal.inria.fr/hal-03716629
Contributor : Emmanuel Menier Connect in order to contact the contributor
Submitted on : Thursday, July 7, 2022 - 4:05:03 PM
Last modification on : Sunday, August 28, 2022 - 5:39:06 PM

Files

example_paper.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03716629, version 1

Citation

Emmanuel Menier, Michele Alessandro Bucci, Mouadh Yagoubi, Lionel Mathelin, Marc Schoenauer. Continuous Methods : Hamiltonian Domain Translation. 2022. ⟨hal-03716629⟩

Share

Metrics

Record views

50

Files downloads

11