Hypothesis formation and language acquisition with an infinitely-often correct teacher
Sanjay K. Jain, Arun Sharma · 1990
The presence of an "infinitely-often correct teacher " in scientific inference and language acquisition is motivated and studied. The treatment is abstract. In the practice of science, a scientist performs experiments to gather experimental data about some phenomenon, and then tries to construct an explanation (or the-ory) for the phenomenon. A model for the practice of science is an inductive inference machine (a scientist) learning a program (an explanation) from the graph (set of exper-iments) of a recursive function (phenomenon). It is argued that this model of science is not an adequate one as scientists, in addition to performing experiments, make use of some approximate explanation (based on the "state of the art") about the phe-nomenon under investigation. An attempt has been made to model this approximate explanation as an additional information in the scientific process. It is shown that inference power of machines is improved in the presence of an approximate explana-tion. The quality of this approximate information is modeled using certain "density" notions. It is shown that additional information about a "better " quality approximate explanation enhances the inference power of learning machines as scientists more than a "not so good " approximate explanation.