Continuous deep learning architectures have recently re-emerged as Neural
Ordinary Differential Equations (Neural ODEs). This infinite-depth approach
theoretically bridges the gap between deep learning and dynamical systems,
offering a novel perspective. However, deciphering the inner working of these
models is still an open challenge, as most applications apply them as generic
black-box modules. In this work we öpen the box", further developing the
continuous-depth formulation with the aim of clarifying the influence of
several design choices on the underlying dynamics.
%0 Generic
%1 massaroli2020dissecting
%A Massaroli, Stefano
%A Poli, Michael
%A Park, Jinkyoo
%A Yamashita, Atsushi
%A Asama, Hajime
%D 2020
%K deeplearning neuralode ode
%T Dissecting Neural ODEs
%U http://arxiv.org/abs/2002.08071
%X Continuous deep learning architectures have recently re-emerged as Neural
Ordinary Differential Equations (Neural ODEs). This infinite-depth approach
theoretically bridges the gap between deep learning and dynamical systems,
offering a novel perspective. However, deciphering the inner working of these
models is still an open challenge, as most applications apply them as generic
black-box modules. In this work we öpen the box", further developing the
continuous-depth formulation with the aim of clarifying the influence of
several design choices on the underlying dynamics.
@misc{massaroli2020dissecting,
abstract = {Continuous deep learning architectures have recently re-emerged as Neural
Ordinary Differential Equations (Neural ODEs). This infinite-depth approach
theoretically bridges the gap between deep learning and dynamical systems,
offering a novel perspective. However, deciphering the inner working of these
models is still an open challenge, as most applications apply them as generic
black-box modules. In this work we "open the box", further developing the
continuous-depth formulation with the aim of clarifying the influence of
several design choices on the underlying dynamics.},
added-at = {2021-09-21T09:12:09.000+0200},
author = {Massaroli, Stefano and Poli, Michael and Park, Jinkyoo and Yamashita, Atsushi and Asama, Hajime},
biburl = {https://www.bibsonomy.org/bibtex/2d9d569e827454b8ea7b9e39b3b71fa0d/annakrause},
description = {2002.08071.pdf},
interhash = {132078e96a914edcaa7367f0f9cf75d4},
intrahash = {d9d569e827454b8ea7b9e39b3b71fa0d},
keywords = {deeplearning neuralode ode},
note = {cite arxiv:2002.08071},
timestamp = {2021-09-21T09:12:09.000+0200},
title = {Dissecting Neural ODEs},
url = {http://arxiv.org/abs/2002.08071},
year = 2020
}