Near-Optimal Multi-Perturbation Experimental Design for Causal Structure Learning.
Scott SussexCaroline UhlerAndreas KrausePublished in: NeurIPS (2021)
Keyphrases
- experimental design
- structure learning
- bayesian networks
- sample size
- markov blanket
- graphical models
- conditional independence
- active learning
- parameter estimation
- empirical studies
- markov logic networks
- feature selection
- transfer learning
- causal models
- markov networks
- maximum likelihood
- probabilistic model
- random variables
- game play
- lower bound
- data sets
- conditional probabilities
- model selection