Learning by creatifying transfer frames

摘要

In the particular kind of learning discussed in this paper, the teacher names a destination and a source. In the sentence, “Robbie is like a fox”, Robbie is the destination and fox is the source. The student, on analyzing the teacher's instruction, computes a filter called a transfer frame. The transfer frame stands between the source and the destination and determines what information is allowed to pass from one to the other.Creating the transfer frame requires two steps: hypothesis and filtering. In the hypothesis step, potentially useful transfer frames are produced through an analysis of the information in the source and its immediate relatives. For a fox, the transfer frames are created through an analysis of the way foxes compare with other small mammals. In the filtering step, the better of the hypothesized frames are selected through a study of the destination frame and its relatives. For Robbie, the robot, the filtering is done by comparing Robbie with other robots.Each time something is learned, the student takes notes on what happened. These notes are used later to justify subsequent conclusions. Also, the notes sometimes make it possible for the student to create his own source-destination pairs by himself.