Learning causal polytrees
The essence of causality can be identified with a graphical structure representing relevance relationships between variables. In this paper the problem of infering causal relations from patterns of dependence is considered. We suppose that there exists a causal model, which is representable by a polytree structure and present an approach to the recovering problem. With this approach we can recover efficiently a polytree structure using marginal and conditional independence tests.
Unable to display preview. Download preview PDF.
- 1.Acid S., Campos L.M. de, Gonzalez A. Molina R., Pérez de la Blanca N.: Learning with CASTLE, in Symbolic and Quantitative approaches to Uncertainty. Lecture Notes in Computer Science 584, R. Kruse R., Siegel P. (Eds). Springer Verlag, (1991) 99–106.Google Scholar
- 2.Acid S., Campos L.M. de, González A. Molina R., Pérez de la Blanca N.: CASTLE: A tool for bayesian learning. Proceedings of the ESPRIT 91 Conference, Commission of the European Communities, (1991) 363–377.Google Scholar
- 3.Campos L.M.de, Huete J.: Independence concepts in Upper and Lower probabilities. Proceedings of the Fourth IPMU Conference, (1992) 129–132.Google Scholar
- 4.Chow C.K., Liu C.N.: Approximating discrete probability distribution with dependence trees. IEEE Transactions on Information Theory 14, (1968) 462–467.Google Scholar
- 5.Cooper G., Herskovits E.: A bayesian method for the induction of Probabilistic Networks from data. Machine Learning 9, (1992) 309–347.Google Scholar
- 6.Geiger D., Paz A., Pearl J.: Learning causal trees from Dependence information. Proceedings of the Eighth National Conference on A.I. (1990) 770–776.Google Scholar
- 7.Geiger D., Paz A., Pearl J.: Learning Simple Causal Structures International Journal of Intelligent Systems, Vol 8, (1993) 231–247.Google Scholar
- 8.Herskovits E., Cooper G.: KUTATO: An Entropy-Driven system for construction of probabilistic expert systems from Databases. Proceedings of the 6th Conference on Uncertainty in Artificial Intelligence, Cambridge MA. (1990) 54–62.Google Scholar
- 9.Pearl J.: Probabilistic Reasoning in intelligent systems: Networks of plausible inference. Morgan and Kaufman, San Mateo (1988).Google Scholar
- 10.Rebane G., Pearl J.: The recovery of causal poly-trees from statistical data, Uncertainty in Artificial Intelligence 3, Kanal L.N., Levitt T.S. and Lemmer J.F. Eds. North-Holland. (1989) 175–182.Google Scholar
- 11.Srinivas S., Russell S., Agogino A.: Automated construction of sparse Bayesian networks from unstructured probabilistic models and domain information. Uncertainty in Artificial Intelligence 5, Henrion M., Shachter R.D., Kanal L.N., Lemmer J.F. (Eds), North-Holland, Amsterdam. (1990) 295–308.Google Scholar