奎克的藏经阁 · 原书印刷页

智慧的疆界 从图灵机到人工智能 · 第 404 页

PDF 物理页索引 419 · 内容简介

书名
智慧的疆界 从图灵机到人工智能
作者
未记录
印刷页
404
PDF 物理页
419(从 0 开始)
章节
内容简介
UUID
cf76bc49-7e9b-4f42-bee8-3050a9fb9feb
来源类型
original_book
页面 SHA-256
a0c264121430628a63ecb882b01c3a133a7b83daa0b4f1d2acd3b2ea65971bf4

1957 年 在“逻辑理论家”公布之后不久,纽厄尔和司马贺发布了一款更具有普适性的问题解决程序,称为“一般问题解决器”(General ProblemSolver,GPS),希望所有任何已形式化的、具备完全信息的问题都可以用这个程序去解决,他们给 GPS 设定的适用范围涉及逻辑推理、定理证明、国际象棋对抗等多个领域。

REPORT ON A GENERAL PROBLEM-SOLVING PROGRAM A. Newell and J. C. Shaw, The RAND Corporation H. A. Simon, Carnegie Institute of Technology United States of America

This paper deals with the theory of problem solving. It describes a program for a digital computer, called General Problem Solver I (GPS), which is part of an investigation into the extremely complex processes that are involved in intelligent, adaptive, and creative behavior. Our principal means of investigation is synthesis: programming large digital computers to exhibit intelligent behavior, studying the structure of these computer programs, and examining the problem-solving and other adaptive behaviors that the programs produce.

A problem exists whenever a problem solver desires some outcome or state of affairs that he does not immediately know how to attain. Imperfect knowledge about how to proceed is at the core of the genuinely problematic. Of course, some initial information is always available. A genuine problemsolving process involves the repeated use of available information to initiate exploration, which discloses, in turn, more information until a way to attain the solution is finally discovered.

Many kinds of information can aid in solving problems: information may suggest the order in which possible solutions should be examined; it may rule out a whole class of solutions previously thought possible; it may provide a cheap test to distinguish likely from unlikely possibilities; and so on. All these kinds of information are heuristics--things that aid discovery. Heuristics seldom provide infallible guidance; they give practical knowledge, possessing only empirical validity. Often they "work," but the results are variable and success is seldom guaranteed.

The theory of problem solving is concerned with discovering and understanding systems of heuristics. What kinds are there? How do very general injunctions ("Draw a figure" or "Simplify") exert their effects? What heuristics do humans actually use? How are new heuristics discovered? And so on. GPS, the program described in this paper, contributes to the theory of problem solving by embodying two very general systems of heuristics--means-ends analysis and planning -- within an organization that allows them to be applied t varying subject matters.

404

附录 人工智能历史大事记