Research map / Probabilistic and scientific modeling / Variational and generative sampling
Scalable Structure Learning of Continuous-Time Bayesian Networks from Incomplete Data
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 gradient-based CTBN structure learning instead of the strongest listed baseline | IRMA switch on · IRMA switch on | AUROC: 0.79 -> 0.88 | Helped |
| Use gradient-based CTBN structure learning instead of the strongest listed baseline | IRMA switch on · IRMA switch on | AUPR: 0.70 -> 0.86 | Helped |
| Use gradient-based CTBN structure learning instead of the strongest listed baseline | IRMA switch off · IRMA switch off | AUROC: 0.86 -> 0.93 | Helped |
| Use gradient-based CTBN structure learning instead of the strongest listed baseline | IRMA switch off · IRMA switch off | AUPR: 0.72 -> 0.92 | Helped |
| Restrict candidate parent sets to at most K=2 (greedy search rather than exhaustive search) (ablation) | CTBN structure learning · IRMA switch on | AUROC: 0.81 -> 0.88 | Helped |
| Restrict candidate parent sets to at most K=2 (greedy search rather than exhaustive search) (ablation) | CTBN structure learning · IRMA switch on | AUPR: 0.86 -> 0.85 | Hurt |
| Restrict candidate parent sets to at most K=2 (greedy search rather than exhaustive search) (ablation) | CTBN structure learning · IRMA switch off | AUROC: 0.93 -> 0.91 | Hurt |
| Restrict candidate parent sets to at most K=2 (greedy search rather than exhaustive search) (ablation) | CTBN structure learning · IRMA switch off | AUPR: 0.92 -> 0.89 | Hurt |
| Initialize mixture weights with the heuristic rather than random values (ablation) | CTBN mixture-weight optimization · Synthetic data, 40 trajectories | AUROC | Helped |
| Approximate the geometric mean by the arithmetic mean for greedy search (ablation) | CTBN greedy structure search · Synthetic incomplete data | AUROC | Hurt |
Closest papers by meaning
- Cluster Variational Approximations for Structure Learning of Continuous-Time Bayesian Networks from Incomplete Data (NeurIPS 2018 · similarity 0.94)
- Marginalized Continuous Time Bayesian Networks for Network Reconstruction from Incomplete Observations (AAAI 2016 · similarity 0.93)
- Extendable and Iterative Structure Learning Strategy for Bayesian Networks (ICLR 2025 · similarity 0.90)
- Active Learning of Continuous-time Bayesian Networks through Interventions (ICML 2021 · similarity 0.89)
- Sequential Monte Carlo Learning for Time Series Structure Discovery (ICML 2023 · similarity 0.88)
- Neural structure learning with stochastic differential equations (ICLR 2024 · similarity 0.88)
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.