Research map / Probabilistic and scientific modeling / Dynamical systems
Time-Reversal Symmetric ODE Network
In Huh, Eunho Yang, Sung Ju Hwang, Jinwoo Shin
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Add time-reversal symmetry loss | ODEN · Duffing simple oscillator | trajectory MSE: 4.05 $\pm$ 2.66 -> 0.31 $\pm$ 0.19 | Improved |
| Add time-reversal symmetry loss | ODEN · Duffing non-linear oscillator | trajectory MSE: 50.81 $\pm$ 26.80 -> 13.78 $\pm$ 14.86 | Improved |
| Combine Hamiltonian structure with time-reversal symmetry loss | HODEN · Duffing non-linear oscillator | trajectory MSE: 17.40 $\pm$ 17.74 -> 10.85 $\pm$ 12.62 | Improved |
| Add time-reversal symmetry loss | ODEN · Duffing forced non-linear oscillator | trajectory MSE: 39.21 $\pm$ 21.19 -> 6.50 $\pm$ 5.59 | Improved |
| Add time-reversal symmetry loss | ODEN · Duffing damped oscillator | trajectory MSE: 1.28 $\pm$ 0.82 -> 0.12 $\pm$ 0.06 | Improved |
| Use time-dependent regularization strength (lambda = 0.5t) rather than a fixed strength (ablation) | TRS-ODEN · Duffing damped oscillator | trajectory MSE | Improved |
| Add time-reversal symmetry loss | ODEN · Real-world coupled double oscillators | MSE: 1.00 $\pm$ 0.22 -> 0.36 $\pm$ 0.06; 0.37 $\pm$ 0.05 -> 0.15 $\pm$ 0.01 | Improved |
| Add time-reversal symmetry loss | ODEN · Reversible strange attractor | trajectory prediction | Improved |
| Combine time-reversal symmetry loss with Hamiltonian structure | HODEN · Duffing non-linear oscillator | energy MSE: 0.22 $\pm$ 0.17 -> 0.29 $\pm$ 0.18 | No clear effect |
| Use fixed lambda = 1 instead of another tested TRS-ODEN regularization setting (ablation) | TRS-ODEN · Duffing damped oscillator | trajectory and energy MSE | Hurt |
| Replace time-reversal regularization with strict Hamiltonian energy conservation | TRS-ODEN · Real-world coupled double oscillators | MSE: 0.36 $\pm$ 0.06 -> 38.13 $\pm$ 2.16; 0.15 $\pm$ 0.01 -> 32.60 $\pm$ 1.96 | Hurt |
Closest papers by meaning
- Physics-Informed Regularization for Domain-Agnostic Dynamical System Modeling (NeurIPS 2024 · similarity 0.89)
- Discovering Physics Laws of Dynamical Systems via Invariant Function Learning (ICML 2025 · similarity 0.86)
- Symmetry-Informed Governing Equation Discovery (NeurIPS 2024 · similarity 0.86)
- Context-Informed Neural ODEs Unexpectedly Identify Broken Symmetries: Insights from the Poincaré–Hopf Theorem (ICML 2025 · similarity 0.84)
- ControlSynth Neural ODEs: Modeling Dynamical Systems with Guaranteed Convergence (NeurIPS 2024 · similarity 0.84)
- ODEFormer: Symbolic Regression of Dynamical Systems with Transformers (ICLR 2024 · similarity 0.83)
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.
