Research map / Probabilistic and scientific modeling / Dynamical systems
Online Bayesian Experimental Design for Partially Observed Dynamical Systems
Sara Perez-Vieites, Sahel Iqbal, Simo Särkkä, Dominik Baumann
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 |
|---|---|---|---|
| select designs sequentially online instead of random designs | BAD-PODS · Two-group SIR (T=200) | TEIG: 5.621 -> 5.918 | Improved |
| select designs sequentially online instead of static offline optimisation | BAD-PODS · Two-group SIR (t=100) | TEIG: 2.529 -> 3.283 | Improved |
| gradient-based design optimisation instead of oracle grid search | BAD-PODS · Two-group SIR (T=200) | TEIG: 6.043 -> 5.918 | No clear effect |
| sequential designs instead of random designs | BAD-PODS · Moving source (T=50) | TEIG: 1.435 -> 1.733 | Improved |
| sequential designs instead of static BED | BAD-PODS · Moving source (t=20) | TEIG: 0.570 -> 0.688 | Improved |
| sequential designs instead of random designs | BAD-PODS · Ecological growth (T=20) | TEIG: 2.344 -> 2.545 | Improved |
| sequential designs instead of static BED on a short horizon | BAD-PODS · Ecological growth (t=10) | TEIG: 1.571 -> 1.579 | No clear effect |
| use online NPF-based BAD-PODS instead of static BED (runtime) | BAD-PODS · Two-group SIR T=150 | wall-clock runtime (min): 31.05 -> 12.39 | Improved |
| use online NPF-based BAD-PODS instead of static BED (runtime, moving source) | BAD-PODS · Moving source T=20 | wall-clock runtime (min): 628 -> 525 | Improved |
| use online BAD-PODS instead of static BED (memory) | BAD-PODS · Two-group SIR T=150 | GPU memory (GB): 25 GB -> 10 GB | Improved |
Closest papers by meaning
- Nesting Particle Filters for Experimental Design in Dynamical Systems (ICML 2024 · similarity 0.89)
- On Estimating the Gradient of the Expected Information Gain in Bayesian Experimental Design (AAAI 2024 · similarity 0.87)
- Bayesian Experimental Design Via Contrastive Diffusions (ICLR 2025 · similarity 0.86)
- Predicting partially observable dynamical systems via diffusion models with a multiscale inference scheme (NeurIPS 2025 · similarity 0.85)
- Online Variational Sequential Monte Carlo (ICML 2024 · similarity 0.85)
- Maximin Robust Bayesian Experimental Design (NeurIPS 2026 · similarity 0.85)
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.
