Research map / Probabilistic and scientific modeling / Dynamical systems
Two-Parameter Flows for Learning Population Dynamics of Physical Systems
Paul Schwerdtner, Tobias Blickhan, Benjamin Peherstorfer
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 |
|---|---|---|---|
| use two-parameter flow instead of gradient-ansatz DICE | population dynamics · Vlasov-Poisson two-stream | avg W2 error: 4.5e-04 -> 3.2e-04 | Improved |
| use two-parameter flow instead of DICE | population dynamics · Vlasov-Poisson bump-on-tail | avg W2 error: 6.5e-03 -> 3.1e-04 | Improved |
| use two-parameter flow instead of HOAM | population dynamics · Vlasov-Poisson bump-on-tail | avg W2 error: 3.8e+01 -> 3.1e-04 | Improved |
| use two-parameter flow instead of action matching | population dynamics · Vlasov-Poisson two-stream | avg W2 error: 4.6e+00 -> 3.2e-04 | Improved |
| use two-parameter flow instead of JKONet* | population dynamics · Vlasov-Poisson bump-on-tail | final W2 error: 7.5e-03 -> 4.7e-04 | Improved |
| fit u by TPF regression instead of pointwise MSE trajectory fitting | TPF · Kolmogorov flow | energy spectrum exponent (target -3): MSE -3.805 vs TPF -2.880 at step 127 (physics model -2.970) | Improved |
| TPF versus MSE-fitted model | TPF · barotropic flow (128x128) | enstrophy decay | Improved |
| TPF versus numerical physics model | TPF · barotropic flow | inference time | Improved |
| TPF versus physical trajectories | TPF · barotropic flow | kinetic energy: roughly two orders of magnitude lower | Improved |
| replace successive OT couplings with two-parameter flow | TPF · evolving Gaussian mixture | regularity of dynamics | Improved |
| avoid gradient ansatz (allow non-gradient dynamics) | TPF · rotating particle instabilities | accuracy | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- DICE — DiCE: The Infinitely Differentiable Monte Carlo Estimator (ICML 2018) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Parametric model reduction of mean-field and stochastic systems via higher-order action matching (NeurIPS 2024 · similarity 0.91)
- Curly Flow Matching for Learning Non-gradient Field Dynamics (NeurIPS 2025 · similarity 0.88)
- First-Order Trajectory Matching: Fast Ensemble Predictions of Chaotic, Turbulent, Stochastic Systems (NeurIPS 2026 · similarity 0.87)
- NETS: A Non-equilibrium Transport Sampler (ICML 2025 · similarity 0.86)
- Wasserstein residuals: Learning Gradient Flows from Population Dynamics (NeurIPS 2026 · similarity 0.86)
- Generating Physical Dynamics under Priors (ICLR 2025 · similarity 0.86)
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.
