Bane was developer in summer term 2003 als preparation and example of my presentation on “statistical methods in artificial intelligence”, my part of an advanced seminar on artificial intelligence at the Ludwig-Maximilians-University, Munich (LMU).

The simulator can only handle networks that are polytrees (i.e. that do not contain cycles). For these family of networks there is a known algorithm with linear complexity, which thus was predestined for this presentation (an implementation was by no means required, it was a voluntary project to obtain a deeper understanding and to be able to explain it better in the presentation).

Handout (german only), slides (german only), Screenshot and full sourcecode (C++, requires GTKmm-2.4 sigc++-2.0 and libxml++-2.6), minimal sourcecode (for study and manual compilation)

This program and the documents do not claim to be complete or correct. Nor that they are an excellent source for understanding AI. People interested in these topics should probably read David Heckermans Tutorials or e.g. these lecture notes of the McGill University, Montreal. Or a good book on artificial intelligence.