AI Paper Atlas

Research map / Time series / Causal and generative time series

MALI: A memory efficient and reverse accurate integrator for Neural ODEs

Juntang Zhuang, Nicha C Dvornek, sekhar tatikonda, James s Duncan

ICLR 2021 · Time series › Causal and generative time series · approach: Neural differential equations · 15 citations (OpenAlex)

Read the paper

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use MALI to train the Neural ODE, then evaluate with RK4Neural ODE18 · ImageNet validationtop-1 accuracy: 70.09 -> 70.21Improved
Replace adjoint training with MALINeural ODE18 · ImageNet validationtop-1 accuracy: 63% -> 70%Improved
Replace ACA training with MALILatent-ODE · Mujoco test, 10% training dataMSE: 0.31 -> 0.35Hurt
Replace ACA training with MALILatent-ODE · Mujoco test, 20% training dataMSE: 0.27 -> 0.27No clear effect
Replace ACA training with MALINeural CDE · Speech Commandsaccuracy: 93.2±0.2 -> 93.7±0.3Improved
Train FFJORD with MALI instead of SemiNormFFJORD · MNIST validationbits per dimension: 0.96 -> 0.87Improved
Train FFJORD with MALI instead of Flow++FFJORD · CIFAR10 validationbits per dimension: 3.28 -> 3.27Improved
Train FFJORD with MALI instead of Residual FlowFFJORD · ImageNet64 validationbits per dimension: 3.76 -> 3.71Improved
Set damping coefficient η to 0.95 instead of 1.0 (ablation)Neural CDE · Speech Commandsaccuracy: 93.7 ± 0.3 -> 93.7 ± 0.1No clear effect
Set damping coefficient η to 0.9 instead of 1.0 (ablation)Neural CDE · Speech Commandsaccuracy: 93.7 ± 0.3 -> 93.5 ± 0.2No clear effect
Set damping coefficient η to 0.85 instead of 1.0 (ablation)Neural CDE · Speech Commandsaccuracy: 93.7 ± 0.3 -> 93.7 ± 0.3No clear effect
Set damping coefficient η to 0.95 instead of 1.0 (ablation)Latent-ODE · Mujoco test, 10% training dataMSE: 0.35 -> 0.36Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy