Title of article :
On the incompatibility of faithfulness and monotone DAG faithfulness Original Research Article
Author/Authors :
David Maxwell Chickering، نويسنده , , Christopher Meek، نويسنده ,
Issue Information :
روزنامه با شماره پیاپی سال 2006
Pages :
14
From page :
653
To page :
666
Abstract :
Cheng, Greiner, Kelly, Bell and Liu [Artificial Intelligence 137 (2002) 43–90] describe an algorithm for learning Bayesian networks that—in a domain consisting of n variables—identifies the optimal solution using image calls to a mutual-information oracle. This result relies on (1) the standard assumption that the generative distribution is Markov and faithful to some directed acyclic graph (DAG), and (2) a new assumption about the generative distribution that the authors call monotone DAG faithfulness (MDF). The MDF assumption rests on an intuitive connection between active paths in a Bayesian-network structure and the mutual information among variables. The assumption states that the (conditional) mutual information between a pair of variables is a monotonic function of the set of active paths between those variables; the more active paths between the variables the higher the mutual information. In this paper, we demonstrate the unfortunate result that, for any realistic learning scenario, the monotone DAG faithfulness assumption is incompatible with the faithfulness assumption. Furthermore, for the class of Bayesian-network structures for which the two assumptions are compatible, we can learn the optimal solution using standard approaches that require only O(image) calls to an independence oracle.
Keywords :
Bayesian networks , Grafical models , Learning , Structure search , Complexity
Journal title :
Artificial Intelligence
Serial Year :
2006
Journal title :
Artificial Intelligence
Record number :
1207483
Link To Document :
بازگشت