ScoreBasedBase<T>
Base Classes · AiDotNet.CausalDiscovery.ScoreBased
Base class for score-based causal discovery algorithms (GES, FGES, Hill Climbing, Tabu, etc.).
Inheritance: CausalDiscoveryBase<T> → ScoreBasedBase<T>
Known derived types: BOSSAlgorithm<T>, ExactSearchAlgorithm<T>, FGESAlgorithm<T>, GESAlgorithm<T>, GRaSPAlgorithm<T>, HillClimbingAlgorithm<T>, K2Algorithm<T>, TabuSearchAlgorithm<T>
For Beginners
Score-based methods give each possible causal graph a “grade” (score) based on how well it explains the data. They then search for the graph with the best grade. Higher scores mean the graph fits the data better while remaining simple (not too many edges).
How It Works
Score-based methods search over the space of DAGs by evaluating each candidate graph using a scoring criterion (typically BIC or BDeu). They use search strategies like greedy equivalence search, hill climbing, or tabu search to find high-scoring graphs.
Properties
| Property | Summary |
|---|---|
CausalDiscoveryCategory Category { get; } |