Friedberg Finds Computer Shows Ability to 'Learn'

Experimental evidence that modern electronic computers may be able to "learn" from experience and devise their own method for solving given problems has been found by Richard M. Friedberg '56, a second-year student at the Medical School.

Friedberg will publish his findings in the January issue of "The IBM Journal of Research and Development." He performed the tests at the IBM Research Center in Poughkeepsie last summer on the 704 Computer.

"If we are ever to make a machine that will speak, understand or translate human language, solve mathematical problems with imagination, practice a profession or direct an organization, either we must reduce these activities to a science so exact that we can tell a machine precisely how to go about doing them or we must develop a machine that can do things without being told precisely how," Friedberg stated.

Finds Computer "Learns"

In the actual experiment, Friedberg simulated the arrangement of his own theoretical "Herman" computer connected to a "teacher" and a "learner" on the IBM 704 computer.

The experiment was conducted to find out whether "Herman-Teacher-Learner" would have to go through an astronomical number of methods or whether the machine had some judgment. The computer was found to "learn" from experience.

"Success was achieved on the problems presented to the machine," Friedberg said, "but I am not satisfied by the way it came about." He noted that he is as close to achieving the ultimate goals of this research "as the first man who discovered that paper sticks to a comb was to discovering radio."