AI Paper Atlas

Research map / Probabilistic and scientific modeling / Dynamical systems

Which priors matter? Benchmarking models for learning latent dynamics

Aleksandar Botev, Andrew Jaegle, Peter Wirnsberger, Daniel Hennes, Irina Higgins

NeurIPS 2021 · Probabilistic and scientific modeling › Dynamical systems · approach: Latent dynamics models · 4 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

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
model dynamics with Neural ODE with time-reversal constraint instead of Hamiltonian priorHGN · 17 datasetsforward VPT best-in-dataset count: 11/17Improved
model dynamics with Neural ODE with time-reversal constraint instead of Hamiltonian priorHGN · 17 datasetsbackward VPT best-in-dataset count: 13/17Improved
model dynamics continuously (Neural ODE instead of discrete RGN Res) (ablation)RGN Res · 17 datasetsforward VPT average rank: 3.0 -> 2.88Improved
use Hamiltonian prior instead of Lagrangian prior (ablation)LGN NEW · 17 datasetsbackward VPT average rank: 2.12 -> 2.59Hurt
use Lagrangian prior instead of Hamiltonian prior (ablation)HGN · 17 datasetsforward VPT average rank: 4.47 -> 4.18Mixed
Lagrangian prior needs matrix inversion for acceleration (ablation)LGN · training costper-step computeHurt
predict residual state updates (RGN Res instead of RGN) (ablation)RGN · 17 datasetsforward VPT average rank: 5.65 -> 3.0Improved
residual prediction (RGN Res instead of autoregressive AR) (ablation)AR · 17 datasetsforward VPT average rank: 5.65 -> 3.0Improved
add time-reversal constraint to Neural ODE (ODE[TR] vs ODE) (ablation)ODE · 17 datasetsbackward VPT average rank: 3.47 -> 1.41Improved
add time-reversal constraint to Neural ODE (ODE[TR] vs ODE) (ablation)ODE · 17 datasetsforward VPT average rank: 2.88 -> 2.47Improved
add friction and control to mass-spring (+c +f vs plain) (ablation)HGN · Mass-springforward VPT: 1.00(0.00) -> 0.41(0.17)Hurt
increase MD system size from 4 to 16 particles (ablation)all models · MDforward VPT: 0.28(0.05) -> 0.00(0.00)Hurt

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