Mining various dependence structures from data are important to many data mining applications. In this paper, several major dependence structure mining tasks are overviewed from statistical learning perspective, with a number of major results on unsupervised learning models that range from a single-object world to a multi-object world. Moreover, efforts towards a key challenge to learning have been discussed in three typical streams, based on generalization error bounds, Ockham principle, and BYY harmony learning, respectively.
CITATION STYLE
Xu, L. (2002). Mining dependence structures from statistical learning perspective. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 2412, pp. 285–306). Springer Verlag. https://doi.org/10.1007/3-540-45675-9_47
Mendeley helps you to discover research relevant for your work.