A reply to Cohen's book review of Creating a Memory of Causal Relationships

Abstract

Progress in machine learning must consist of periods of exploration followed by periods of more thorough careful investigation of issues raised during exploration. The research reported in Creating a Memory of Causal Relationships is exploratory in that it addresses a problem that was not previously investigated in the mainstream of machine learning research. However, I feel that the problem studied was worthy of investigation and is worthy of continued investigation since it corresponds to an important part of the human learning process. © 1993 Kluwer Academic Publishers.