View source for Heckerman, JMLR 2000

From Cohen Courses

← Heckerman, JMLR 2000

You do not have permission to edit this page, for the following reason:

---

You can view and copy the source of this page.

== Citation == Dependency Networks for Inference, Collaborative Filtering, and Data Visualization. David Heckerman, David Maxwell Chickering, Christopher Meek, Robert Rounthwaite, Carl Kadie; in JMLR, 1(Oct):49-75, 2000 == Online version == [http://jmlr.csail.mit.edu/papers/volume1/heckerman00a/heckerman00a.pdf] == Summary == In this [[Category::paper]], author describe a graphical model for probabilistic relationship, an alternative to the bayesian network, called dependency network. The dependency network, unlike bayesian network is potentially cyclic. The dependency network are well suited to task of predicting preferences like in collaborative filtering. The dependency network is not good for encoding causal relationship. == Brief description of the method == * Consistent Dependency Network Given a domain of interest having set of finite variables <math>X = (X_1, ... , X_n)</math> with a positive joint distribution <math>p(x)</math>, a consistent dependency network for <math>X</math> is a pair <math>(G, P)</math> where <math>G</math> is a cyclic directed graph and <math>P</math> is a set of conditional probability distributions. The parents of node <math>X_i</math>, denoted <math>Pa_i</math>, correspond to those variable <math>Pa_i \subseteq (X_1, .. , X_{i-1}, X_{i+1},... , X_n)</math> that satisfy: <math>p(x_i|pa_i) = p(x_i | x_1,...., x_{i-1}, x_{i+1}, ..., x_n) = p(x_i | x \setminus x_i)</math> * General Dependency Network == Experimental Result == == Related papers ==

Return to Heckerman, JMLR 2000.