Abstract

In this paper we address the problem of learning the structure in nonlinear Markov networks with continuousvariables. Markov networks are well suited to model relationships which do not exhibit a natural causal ordering. We use neural network structures to model the quantitative relationships between variables. Using two data sets we show that interesting structures can be found using our approach. 1 Introduction Knowledge about independence or conditional independence between variables is most helpful in "understanding" a domain. An intuitive representation of independencies is achieved by graphical stochastical models in which independency statements can be extracted from the structure of the graph. The two most popular types of graphical stochastical models are Bayesian networks which use a directed graph, and Markov networks which use an undirected graph. Whereas Bayesian networks are well suited to represent causal relationships, Markov networks are mostly used in cases where the...

Citations