| "Some Studies in Machine Learning Using the Game of Checkers", Arthur L. Samuel, IBM Journal of Research and Development 3, July 1959, 211--229, also in [Feigenbaum and Feldman 63]. |
....things such as search, and representations based on individuals, much easier to program. At the time, most programs were written in assembly language. It was a tedious job to write search procedures, especially recursive procedures in the machine languages of the day, although some people such as [Samuel 59] another Dartmouth participant) were spectacularly successful. Newell and Simon owed much of their success in developing the Logic Theorist and their later General Problem Solver [Newell, Shaw and Simon 59] to their use of an interpreted language (IPLV see [Newell, Shaw and Simon 61] which ....
....it is wise to rely so heavily on search, as bigger problems will have exponentially bad effects on search time in fact [Newell, Shaw and Simon 58] argue just this, but produced a markedly slower chess program because of the complexity of static evaluation and search control. Some, such as [Samuel 59] with his checker s playing program, did worry about keeping things on a human timescale. Slagle 63] in his symbolic integration program, was worried about being economically competitive with humans, but as he points out in the last two paragraphs of his paper, the explosive increase in ....
"Some Studies in Machine Learning Using the Game of Checkers", Arthur L. Samuel, IBM Journal of Research and Development 3, July 1959, 211--229, also in [Feigenbaum and Feldman 63].
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC