Identifying Cellular Automata Rules
Ken-Ichi Maeda and Chiaki Sakama
This paper studies a method for identifying cellular automata rules (CA rules). Given a sequence of CA configurations, we first seek an appropriate neighborhood of a cell and collect cellular changes of states as evidences. The collected evidences are then classified using a decision tree, which is used for constructing CA transition rules. Conditions for classifying evidences in a decision tree are computed using genetic programming. We perform experiments using several types of CAs and verify that the proposed method successfully identifies correct CA rules.