Research map / Learning theory and algorithms / Sample complexity
Active Structure Learning of Bayesian Networks in an Observational Setting
Noa Ben-David, Sivan Sabato
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 |
|---|---|---|---|
| use ActiveBNSL instead of the naive algorithm at largest sample budget r=2^19 | naive structure learning · Sachs network | samples used (ratio): 776x10^16 -> 49x10^16 (6%) | Improved |
| use ActiveBNSL instead of naive algorithm on Asia at r=2^19 | naive structure learning · Asia network | samples used (ratio): 1055x10^15 -> 94x10^15 (8%) | Improved |
| use ActiveBNSL on synthetic d=12, r=2^19 | naive structure learning · Synthetic d=12 | samples used (ratio): 475x10^16 -> 74x10^16 (15%) | Improved |
| use ActiveBNSL at small sample budget (r=2^11) on benchmark networks | naive structure learning · Cancer network | samples used (ratio): 96x10^10 -> 116x10^10 (120%) | Hurt |
| use ActiveBNSL on small synthetic d=6 networks | naive structure learning · Synthetic d=6 | samples used (ratio): 43x10^12 -> 51x10^12 (118%) | Hurt |
| use ActiveBNSL on synthetic d=9, r=2^19 | naive structure learning · Synthetic d=9 | samples used (ratio): 165x10^16 -> 30x10^16 (18%) | Improved |
| use ActiveBNSL on synthetic d=7, r=2^19 | naive structure learning · Synthetic d=7 | samples used (ratio): 62x10^16 -> 21x10^16 (34%) | Improved |
| use ActiveBNSL on Survey network at r=2^13 | naive structure learning · Survey network | samples used (ratio): 43x10^12 -> 21x10^12 (48%) | Improved |
Closest papers by meaning
- Structure Learning in Bayesian Networks of a Moderate Size by Efficient Sampling (JMLR 2016 · similarity 0.91)
- Extendable and Iterative Structure Learning Strategy for Bayesian Networks (ICLR 2025 · similarity 0.91)
- Learning Identifiable Gaussian Bayesian Networks in Polynomial Time and Sample Complexity (NeurIPS 2017 · similarity 0.89)
- Learning Bayesian Networks in the Presence of Structural Side Information (AAAI 2022 · similarity 0.89)
- Improving Bayesian Network Structure Learning in the Presence of Measurement Error (JMLR 2022 · similarity 0.89)
- Transductive Active Learning: Theory and Applications (NeurIPS 2024 · similarity 0.87)
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.
