Experience, Memory, and Reasoning

Experience, Memory, and Reasoning

Experience, Memory, and Reasoning

Experience, Memory, and Reasoning

Excerpt

The chapters in this volume, though different in topic, style, and even scientific discipline, all share a common theme: that knowledge is first and foremost the product of experience. This obvious truth about human knowledge has significant ramifications for artificial intelligence (AI) and cognitive science. It means that to understand human knowledge and how it is organized and applied, we have to understand how it was acquired. It means that to claim that a representation of knowledge is adequate, we have to show how that knowledge could be learned and how it could be updated in the face of new experience.

KNOWLEDGE AND AI

The importance of knowledge for intelligence is one of AI's few consensual beliefs. Knowledge, in large quantities organized into usable chunks, is an essential ingredient of a great deal of intelligent behavior. How to chunk knowledge about the world to make it usable for reasoning is an unavoidable problem in AI research. The real world is full of objects and tableaux that have no identifying labels to help the visitor. To understand the world, we must carry around in our heads a field guide, a guide so large that the quality of its index is a major factor in the usefulness of the guide.

The metaphor is incomplete, however, until we make one further point: the guide is really a notebook that each of us writes and maintains. We are born with pencil and paper and have to take it from there. As a result, we each have a different field guide, in our own handwriting, with illegible scribbles, mistakes, corrections, and redundancies, and at no point can we stop and say that the book is complete.

Author Advanced search

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.