We introduce a new family of deep neural network models. Instead of
specifying a discrete sequence of hidden layers, we parameterize the derivative
of the hidden state using a neural network. The output of the network is
computed using a black-box differential equation solver. These continuous-depth
models have constant memory cost, adapt their evaluation strategy to each
input, and can explicitly trade numerical precision for speed. We demonstrate
these properties in continuous-depth residual networks and continuous-time
latent variable models. We also construct continuous normalizing flows, a
generative model that can train by maximum likelihood, without partitioning or
ordering the data dimensions. For training, we show how to scalably
backpropagate through any ODE solver, without access to its internal
operations. This allows end-to-end training of ODEs within larger models.
%0 Unpublished Work
%1 ChenEtal2018arxiv
%A Chen, Ricky T. Q.
%A Rubanova, Yulia
%A Bettencourt, Jesse
%A Duvenaud, David
%D 2018
%K ANN ODE
%T Neural Ordinary Differential Equations
%U http://arxiv.org/abs/1806.07366
%X We introduce a new family of deep neural network models. Instead of
specifying a discrete sequence of hidden layers, we parameterize the derivative
of the hidden state using a neural network. The output of the network is
computed using a black-box differential equation solver. These continuous-depth
models have constant memory cost, adapt their evaluation strategy to each
input, and can explicitly trade numerical precision for speed. We demonstrate
these properties in continuous-depth residual networks and continuous-time
latent variable models. We also construct continuous normalizing flows, a
generative model that can train by maximum likelihood, without partitioning or
ordering the data dimensions. For training, we show how to scalably
backpropagate through any ODE solver, without access to its internal
operations. This allows end-to-end training of ODEs within larger models.
@unpublished{ChenEtal2018arxiv,
abstract = {We introduce a new family of deep neural network models. Instead of
specifying a discrete sequence of hidden layers, we parameterize the derivative
of the hidden state using a neural network. The output of the network is
computed using a black-box differential equation solver. These continuous-depth
models have constant memory cost, adapt their evaluation strategy to each
input, and can explicitly trade numerical precision for speed. We demonstrate
these properties in continuous-depth residual networks and continuous-time
latent variable models. We also construct continuous normalizing flows, a
generative model that can train by maximum likelihood, without partitioning or
ordering the data dimensions. For training, we show how to scalably
backpropagate through any ODE solver, without access to its internal
operations. This allows end-to-end training of ODEs within larger models.},
added-at = {2020-04-25T19:07:37.000+0200},
author = {Chen, Ricky T. Q. and Rubanova, Yulia and Bettencourt, Jesse and Duvenaud, David},
biburl = {https://www.bibsonomy.org/bibtex/2d1fc1274c5416b5631ada1f3dd3e0845/johannrudi},
description = {Neural Ordinary Differential Equations},
interhash = {266f336a7ba7b004150cccc49cd576f6},
intrahash = {d1fc1274c5416b5631ada1f3dd3e0845},
keywords = {ANN ODE},
note = {cite arxiv:1806.07366},
timestamp = {2020-04-25T19:07:37.000+0200},
title = {Neural Ordinary Differential Equations},
url = {http://arxiv.org/abs/1806.07366},
year = 2018
}