Process and Policy: Resource-Bounded NonDemonstrative Reasoning

作者: Ronald P. Loui

DOI: 10.1111/0824-7935.00055

关键词: Reasoning systemCase-based reasoningMathematicsDeductive reasoningCausal reasoningEpistemologyQualitative reasoningDefeasible reasoningArtificial intelligenceAnalytic reasoningNon-monotonic logic

摘要: This paper investigates the appropriateness of formal dialectics as a basis for nonmonotonic reasoning and defeasible that takes computational limits seriously. Rules can come into conflict should be regarded policies, which are inputs to deliberative processes. Dialectical protocols appropriate such deliberations when resources bounded search is serial. AI, it claimed here, now perfectly positioned correct many misconceptions about have resulted from mathematical logic's enormous success in this century: among them, (1) all reasons demonstrative, (2) rational belief constrained, not constructed, (3) process disputation essential reasoning. AI mainly provides new impetus formalize alternative (but older) conception reasoning, mechanisms with create compelling formalism describes control processes. The technical contributions here are: partial justification dialectic based on controlling search; observation subsumed under certain kinds dialectics; portrayal inference knowledge bases policy reasoning; review logics dialogue proposed extensions; preformal initial discussion aspects variations dialectical systems nondemonstrative reasons.

参考文章(0)