Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Modulated Neural ODEs

About

Neural ordinary differential equations (NODEs) have been proven useful for learning non-linear dynamics of arbitrary trajectories. However, current NODE methods capture variations across trajectories only via the initial state value or by auto-regressive encoder updates. In this work, we introduce Modulated Neural ODEs (MoNODEs), a novel framework that sets apart dynamics states from underlying static factors of variation and improves the existing NODE methods. In particular, we introduce $\textit{time-invariant modulator variables}$ that are learned from the data. We incorporate our proposed framework into four existing NODE variants. We test MoNODE on oscillating systems, videos and human walking trajectories, where each trajectory has trajectory-specific modulation. Our framework consistently improves the existing model ability to generalize to new dynamic parameterizations and to perform far-horizon forecasting. In addition, we verify that the proposed modulator variables are informative of the true unknown factors of variation as measured by $R^2$ scores.

Ilze Amanda Auzina, \c{C}a\u{g}atay Y{\i}ld{\i}z, Sara Magliacane, Matthias Bethge, Efstratios Gavves• 2023

Related benchmarks

TaskDatasetResultRank
Future state predictionBOUNCING BALL (test)
MSE0.0164
8
Trajectory ForecastingSinusoidal data T=50 (test)
Test MSE0.04
6
Trajectory ForecastingSinusoidal data T=150 (test)
Test MSE0.29
6
Trajectory ForecastingPredator-prey data T=100 (test)
Test MSE0.74
6
Trajectory ForecastingPredator-prey data T=300 (test)
Test MSE4.33
6
Future state predictionCMU Mocap (test)
MSE (Test Set)57.7
3
Factor of Variation PredictionSine (test)
R2 Score0.99
2
Factor of Variation PredictionPredator-Prey (PP) (test)
R20.39
2
Factor of Variation PredictionBouncing Ball (BB) (test)
R2 Score0.58
2
Future state predictionROT.MNIST (test)
MSE0.03
2
Showing 10 of 11 rows

Other info

Code

Follow for update