A reasoner can predict a device’s display without knowing how it relates to the physical state. It can find a successful repair without retaining the evidence needed to justify it. It can store a correct proof whose premises no longer describe the system on which it is about to act. This book asks what a representation must retain for knowledge to remain usable.
Machine-Native ReasoningReading edition
Preface
Reading tools
Reading library
Text settings
Saved in this browser for reading chapters. Browser zoom also works.
Loading chapter…