Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Constrained Bayesian Experimental Design via Online Planning
Yujia Guo, Daolang Huang, Xinyu Zhang, Sammie Katt, Samuel Kaski, Ayush Bharti
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 lookahead planning (H = 1) to myopic COPEx (ablation) | COPEx · CES task, budget 100 | EIG: 5.26 -> 7.03 | Improved |
| add lookahead planning (H = 1) to myopic COPEx (ablation) | COPEx · CES task, budget 150 | EIG: 6.12 -> 7.47 | Improved |
| increase planning horizon from H = 1 to H = 3 (ablation) | COPEx · CES task, budget 100 | EIG: 7.03 -> 6.36 | Hurt |
| warm-start scenario tree with amortized policy (one tree) instead of random multi-start (10 trees) (ablation) | COPEx · location finding, H = 1 | cumulative EIG and runtime | Improved |
| increase planning horizon H (ablation) | COPEx · location finding | EIG | Mixed |
| increase number of branches m per node (ablation) | COPEx · location finding | EIG | No clear effect |
| use constraint-aware COPEx planning instead of VPCE and post-hoc masked ALINE | constrained BED baselines · location finding, delta in {0.05, 0.1, 0.2} | final EIG | Improved |
| add non-myopic lookahead (COPEx-1, COPEx-2) over myopic COPEx (ablation) | COPEx · cost-aware active learning with transition constraint | RMSE vs total cost | Improved |
| add longer horizon (COPEx-3) under small budget (ablation) | COPEx · cost-aware active learning, small budget | RMSE | Improved |
| add longer horizon under large budget (ablation) | COPEx · cost-aware active learning, large budget | RMSE | No clear effect |
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.
- ALINE — ALINE: Joint Amortization for Bayesian Inference and Active Data Acquisition (NeurIPS 2025) · better on 2 of 3 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
- Step-DAD: Semi-Amortized Policy-Based Bayesian Experimental Design (ICML 2025 · similarity 0.89)
- Amortized Bayesian Experimental Design for Decision-Making (NeurIPS 2024 · similarity 0.88)
- Nesting Particle Filters for Experimental Design in Dynamical Systems (ICML 2024 · similarity 0.87)
- On Estimating the Gradient of the Expected Information Gain in Bayesian Experimental Design (AAAI 2024 · similarity 0.87)
- PASOA- PArticle baSed Bayesian Optimal Adaptive design (ICML 2024 · similarity 0.86)
- Maximin Robust Bayesian Experimental Design (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.
