Research map / Probabilistic and scientific modeling / Differential equations and PDEs
Amortized Reparametrization: Efficient and Scalable Variational Inference for Latent SDEs
Kevin Course, Prasanth B. Nair
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 |
|---|---|---|---|
| Train with amortized reparametrization (ARCTA) instead of a neural ODE with adaptive solver | latent neural ODE/SDE · Lotka-Volterra | model evaluations (NFE) to reach similar validation RMSE | Improved |
| Reduce ODE solver tolerance for NODE (ablation) | neural ODE baseline · Lotka-Volterra | validation RMSE vs NFE | Hurt |
| Use ARCTA gradients instead of adjoint gradients | parameter tuning of Lorenz system · Lorenz, long time series (50 and 100 seconds) | gradient stability (l2 norm of gradient) | Improved |
| Use ARCTA latent SDE on motion capture | latent SDE · MOCAP | test RMSE | No clear effect |
Closest papers by meaning
- Closing the Approximation Gap in Simulation-free Latent SDEs (NeurIPS 2026 · similarity 0.90)
- Latent SDEs on Homogeneous Spaces (NeurIPS 2023 · similarity 0.90)
- SING: SDE Inference via Natural Gradients (NeurIPS 2025 · similarity 0.89)
- Variational Inference for SDEs Driven by Fractional Noise (ICLR 2024 · similarity 0.87)
- Disentangling Continuous-Time Latent Dynamics: Identifiability of Latent SDEs via Diffusion Shifts (NeurIPS 2026 · similarity 0.86)
- Amortized Structured Stochastic Variational Inference for Gaussian Process Latent Variable Models (NeurIPS 2026 · 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.
