RNODE

Reported on 6 benchmarks across 2 tasks · 1 paper · 4 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Methodology4 results

Density EstimationonImageNet 64x64
Log-likelihood· 2020-02-07
3.83
SOTA
How to train your neural ODE: the world of Jacobian and kinetic regularization arXiv:2002.02798
Density EstimationonCelebA-HQ 256x256
Log-likelihood· 2020-02-07
1.04
SOTA
How to train your neural ODE: the world of Jacobian and kinetic regularization arXiv:2002.02798
Density EstimationonMNIST
NLL (bits/dim)· 2020-02-07
0.97
best: 0.134 (Identity)
SOTA
How to train your neural ODE: the world of Jacobian and kinetic regularization arXiv:2002.02798
Density EstimationonCIFAR-10
NLL (bits/dim)· 2020-02-07
3.38
best: 2.42 (i-DODE)
How to train your neural ODE: the world of Jacobian and kinetic regularization arXiv:2002.02798

Medical2 results

Image GenerationonCelebA-HQ 256x256
bits/dimension· 2020-02-07
1.04
SOTA
How to train your neural ODE: the world of Jacobian and kinetic regularization arXiv:2002.02798
Image GenerationonMNIST
bits/dimension· 2020-02-07
0.97
best: 0.65 (Locally Masked PixelCNN (8 orders))
How to train your neural ODE: the world of Jacobian and kinetic regularization arXiv:2002.02798