Salīdzināt metodes

	Algoritmi cēloņsakarību atklāšanai (PC, FCI, LiNGAM)	DBSCAN	Grafu uzmanības tīkls (Graph Attention Network, GAT)
Nozare	Cēloņsakarību secināšana	Mašīnmācīšanās	Dziļā mācīšanās
Saime	Regression model	Machine learning	Machine learning
Izcelsmes gads	2000	1996	2018
Autors	Spirtes, Glymour & Scheines (PC/FCI); Shimizu et al. (LiNGAM)	Ester, M., Kriegel, H.-P., Sander, J. & Xu, X.	Veličković, P. et al.
Tips	Causal structure learning	Density-based clustering algorithm	Graph neural network (attention-based)
Pirmavots	Spirtes, P., Glymour, C., & Scheines, R. (2000). Causation, Prediction, and Search (2nd ed.). MIT Press. ISBN: 978-0262194402	Ester, M., Kriegel, H.-P., Sander, J. & Xu, X. (1996). A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise. Proceedings of the 2nd KDD, 226–231.	Veličković, P. et al. (2018). Graph Attention Networks. ICLR.
Citi nosaukumi	PC algorithm, FCI algorithm, LiNGAM, causal structure learning	DBSCAN Kümeleme, density-based clustering, density-based spatial clustering	Graf Dikkat Ağı (GAT), GAT, graph attention network, attention-based graph neural network
Saistītās	5	3	4
Kopsavilkums	Causal discovery is a family of algorithms that automatically learn a directed acyclic graph (DAG) describing causal structure directly from observational data. The constraint-based PC and FCI algorithms were developed by Spirtes, Glymour and Scheines (2000), while the LiNGAM model of Shimizu et al. (2006) exploits linear non-Gaussian structure to orient edges.	DBSCAN is a density-based clustering algorithm, introduced by Ester, Kriegel, Sander and Xu in 1996, that groups together points lying in dense regions and flags points in sparse regions as noise. It is effective on noisy data and on clusters of irregular, non-spherical shapes.	The Graph Attention Network (GAT), introduced by Veličković and colleagues in 2018, is a graph neural network variant that learns how much importance to assign to each neighbouring node through a self-attention mechanism. On heterogeneous neighbourhoods and relational classification it produces results superior to graph convolutional networks (GCN).