Multivariate Time Series Forecasting on MuJoCo

Metric: MSE (10^-2, 50% missing) (lower is better)

LeaderboardDataset

Loading chart...

Results

Sort:

#	Model↕	MSE (10^-2, 50% missing)▲	Extra Data	Paper	Date↕	Code
1	Latent ODE (ODE enc)	1.258	No	Latent ODEs for Irregularly-Sampled Time Series	2019-07-08	Code
2	Latent ODE (RNN enc.)	1.377	No	Neural Ordinary Differential Equations	2018-06-19	Code
3	RNN-VAE	1.782	No	Neural Ordinary Differential Equations	2018-06-19	Code
4	RNN GRU-D	5.833	No	Recurrent Neural Networks for Multivariate Time ...	2016-06-06	Code
5	ODE-RNN	26.463	No	Latent ODEs for Irregularly-Sampled Time Series	2019-07-08	Code
6	RNN ∆t	30.571	No	-	-	-

#1Latent ODE (ODE enc)SOTA
1.258
MSE (10^-2, 50% missing)· 2019-07-08
Latent ODEs for Irregularly-Sampled Time Series Code
#2Latent ODE (RNN enc.)SOTA
1.377
MSE (10^-2, 50% missing)· 2018-06-19
Neural Ordinary Differential Equations Code
#3RNN-VAESOTA
1.782
MSE (10^-2, 50% missing)· 2018-06-19
Neural Ordinary Differential Equations Code
#4RNN GRU-DSOTA
5.833
MSE (10^-2, 50% missing)· 2016-06-06
Recurrent Neural Networks for Multivariate Time Series with Missing Values Code
#5ODE-RNN
26.463
MSE (10^-2, 50% missing)· 2019-07-08
Latent ODEs for Irregularly-Sampled Time Series Code
#6RNN ∆t
30.571
MSE (10^-2, 50% missing)
No paper