Inference Policies

TL;DR

This paper explores the concept of inference policies in AI systems, emphasizing domain-specific tailoring and challenging the necessity of adherence to general rational inference theories.

Contribution

It introduces the idea of domain-specific inference policies and analyzes standard and nonstandard policies to demonstrate their flexibility and applicability.

Findings

Inference policies can be tailored to specific problem domains.

Bayesian reasoning may lead to non-Bayesian inference procedures.

Standard and nonstandard inference policies exhibit diverse characteristics.

Abstract

It is suggested that an AI inference system should reflect an inference policy that is tailored to the domain of problems to which it is applied -- and furthermore that an inference policy need not conform to any general theory of rational inference or induction. We note, for instance, that Bayesian reasoning about the probabilistic characteristics of an inference domain may result in the specification of an nonBayesian procedure for reasoning within the inference domain. In this paper, the idea of an inference policy is explored in some detail. To support this exploration, the characteristics of some standard and nonstandard inference policies are examined.