Need help with DILATE?
Click the “chat” button below for chat support from the developer who created it, or find similar developers for support.

About the developer

vincent-leguen
157 Stars 33 Forks Other 21 Commits 8 Opened issues

Description

Code for our NeurIPS 2019 paper "Shape and Time Distortion Loss for Training Deep Time Series Forecasting Models"

Services available

!
?

Need anything else?

Contributors list

# 110,977
Haskell
Jupyter...
opencl
pytorch
1 commit

DILATE: DIstortion Loss with shApe and tImE

Vincent Le Guen, Nicolas Thome

Code for our NeurIPS 2019 paper "Shape and Time Distortion Loss for Training Deep Time Series Forecasting Models"

If you find this code useful for your research, please cite our paper:

@incollection{leguen19dilate,
title = {Shape and Time Distortion Loss for Training Deep Time Series Forecasting Models},
author = {Le Guen, Vincent and Thome, Nicolas},
booktitle = {Advances in Neural Information Processing Systems},
pages = {4191--4203},
year = {2019}
}

Abstract

This paper addresses the problem of time series forecasting for non-stationary signals and multiple future steps prediction. To handle this challenging task, we introduce DILATE (DIstortion Loss including shApe and TimE), a new objective function for training deep neural networks. DILATE aims at accurately predicting sudden changes, and explicitly incorporates two terms supporting precise shape and temporal change detection. We introduce a differentiable loss function suitable for training deep neural nets, and provide a custom back-prop implementation for speeding up optimization. We also introduce a variant of DILATE, which provides a smooth generalization of temporally-constrained Dynamic Time Warping (DTW). Experiments carried out on various non-stationary datasets reveal the very good behaviour of DILATE compared to models trained with the standard Mean Squared Error (MSE) loss function, and also to DTW and variants. DILATE is also agnostic to the choice of the model, and we highlight its benefit for training fully connected networks as well as specialized recurrent architectures, showing its capacity to improve over state-of-the-art trajectory forecasting approaches.

We use cookies. If you continue to browse the site, you agree to the use of cookies. For more information on our use of cookies please see our Privacy Policy.