Skip to main content
eScholarship
Open Access Publications from the University of California

Self-correcting generalization

Abstract

A system is described which creates and generalizes rules from examples. The system can recover from an initially misleading input sequence by keeping evidence which supports (or doesn't support) a given generalization. By undoing over-generalizations, the system maintains a minimal set of rules for a given set of inputs.

Main Content
For improved accessibility of PDF content, download the file to your device.
Current View