Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Active Learning of Continuous-time Bayesian Networks through Interventions
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 the variational Box-Hill criterion (VBHC) instead of the classical Box-Hill criterion (BHC) to select interventions | CTBN active learning · Synthetic CTBNs | structure and parameter learning speed | Helped |
| Select interventions using VBHC rather than passive observation | CTBN structure learning · Synthetic CTBNs | AUROC and AUPR | Helped |
| Select interventions using VBHC rather than passive observation | CTBN parameter learning · Synthetic CTBNs | posterior-averaged MSE | Helped |
| Select interventions using VBHC rather than passive observation | CTBN structure learning · British Household Panel Survey | posterior entropy | Helped |
| Use sample estimates of VBHC rather than sample estimates of EIG to recommend interventions | CTBN active learning · Synthetic CTBNs | mutual information between history and recommended intervention | Helped |
| Use the negative VBHC design, which selects the worst possible intervention | CTBN structure learning · British Household Panel Survey | AUROC and AUPR convergence | Hurt |
| Intervene on a slow node that points to a fast node | VBHC intervention selection · Synthetic CTBNs | posterior-averaged MSE | Helped |
| Intervene to set a slow parent to a state that is rarely observed passively | CTBN parameter learning · Synthetic two-node CTBN example | information gain about child-node rates | Helped |
Closest papers by meaning
- Scalable Structure Learning of Continuous-Time Bayesian Networks from Incomplete Data (NeurIPS 2019 · similarity 0.89)
- Cluster Variational Approximations for Structure Learning of Continuous-Time Bayesian Networks from Incomplete Data (NeurIPS 2018 · similarity 0.89)
- Sample Efficient Active Learning of Causal Trees (NeurIPS 2019 · similarity 0.88)
- Learning Continuous-Time Bayesian Networks in Relational Domains: A Non-Parametric Approach (AAAI 2016 · similarity 0.88)
- Marginalized Continuous Time Bayesian Networks for Network Reconstruction from Incomplete Observations (AAAI 2016 · similarity 0.88)
- Intervention and Conditioning in Causal Bayesian Networks (NeurIPS 2024 · 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.