Understanding Intelligence

Front Cover
MIT Press, Jul 27, 2001 - Computers - 700 pages
The book includes all the background material required to understand the principles underlying intelligence, as well as enough detailed information on intelligent robotics and simulated agents so readers can begin experiments and projects on their own.

By the mid-1980s researchers from artificial intelligence, computer science, brain and cognitive science, and psychology realized that the idea of computers as intelligent machines was inappropriate. The brain does not run "programs"; it does something entirely different. But what? Evolutionary theory says that the brain has evolved not to do mathematical proofs but to control our behavior, to ensure our survival. Researchers now agree that intelligence always manifests itself in behavior—thus it is behavior that we must understand. An exciting new field has grown around the study of behavior-based intelligence, also known as embodied cognitive science, "new AI," and "behavior-based AI."

This book provides a systematic introduction to this new way of thinking. After discussing concepts and approaches such as subsumption architecture, Braitenberg vehicles, evolutionary robotics, artificial life, self-organization, and learning, the authors derive a set of principles and a coherent framework for the study of naturally and artificially intelligent systems, or autonomous agents. This framework is based on a synthetic methodology whose goal is understanding by designing and building.

The book includes all the background material required to understand the principles underlying intelligence, as well as enough detailed information on intelligent robotics and simulated agents so readers can begin experiments and projects on their own. The reader is guided through a series of case studies that illustrate the design principles of embodied cognitive science.

 

Contents

The Study of Intelligence
3
11 Characterizing Intelligence
6
The Synthetic Approach
21
Foundations of Classical Artificial Intelligence and Cognitive Science
35
22 The Cognitivistic Paradigm
39
23 An Architecture for an Intelligent Agent
47
The Fundamental Problems of Classical Artificial Intelligence and Cognitive Science
59
32 Some WellKnown Problems with Classical Systems
63
103 Design Principles in Context
318
The Principle of Parallel Loosely Coupled Processes
327
111 Control Architectures for Autonomous Agents
330
112 Traditional Views on Control Architectures
337
113 Parallel Decentralized Approaches
345
A SelfSufficient Garbage Collector
357
The Principle of SensoryMotor Coordination
377
Traditional Approaches
378

33 The Fundamental Problems
64
34 Remedies and Alternatives
74
A Framework for Embodied Cognitive Science
79
Embodied Cognitive Science Basic Concepts
81
41 Complete Autonomous Agents
82
42 Biological and Artificial Agents
99
43 Designing for EmergenceLogicBased and Embodied Systems
111
44 Explaining Behavior
127
Neural Networks for Adaptive Behavior
139
51 From Biological to Artificial Neural Networks
140
52 The Four or Five Basics
143
53 Distributed Adaptive Control
152
54 Types of Neural Networks
167
A Polemic Digression
172
Approaches and Agent Examples
179
Braitenberg Vehicles
181
62 The Fourteen Vehicles
182
63 Segmentation of Behavior and the Extended Braitenberg Architecture
195
The Subsumption Architecture
199
71 BehaviorBased Robotics
201
72 Designing a SubsumptionBased Robot
202
73 Examples of SubsumptionBased Architectures
206
The Subsumption Approach to Designing Intelligent Systems
219
Artificial Evolution and Artificial Life
227
81 Basic Principles
230
Evolving a Neural Controller for an Autonomous Agent
234
83 Examples of Artificially Evolved Agents
240
Cell Growth form GenomeBased CelltoCell Communication
250
85 Real Robots Evolution of Hardware and Simulation
255
Additional Examples
260
87 Methodological Issues and Conclusions
270
Other Approaches
277
92 Behavioral Economics
283
93 SchemaBased Approaches
292
Principles of Intelligent Systems
297
Design Principles of Autonomous Agents
299
102 Design Principles for Autonomous Agents
302
122 The SensoryMotor Coordination Approach
392
The SMC Agents
407
Active Vision
431
The Principles of Cheap Design Redundancy and Ecological Balance
435
132 The Redundancy Principle
446
133 The Principle of Ecological Balance
455
The Value Principle
467
141 Value Systems
469
142 SelfOrganization
475
143 Learning in Autonomous Agents
485
A Case Study
503
152 Problems of Classical Notions of Memory
506
153 The FrameofReference Problem in Memory Research
511
154 Alternatives
516
155 Implications for Memory Research
530
Design and Evaluation
535
Agent Design Considerations
537
161 Preliminary Design Considerations
539
162 Agent Design
542
Control Architectures
562
164 Summary and a Fundamental Issue
569
Evaluation
577
171 The Basics of Agent Evaluation
578
172 Performing Agent Experiments
588
173 Measuring Behavior
593
Future Directions
605
Theory Technology and Applications
607
182 Theory and Technology
612
183 Applications
618
Intelligence Revisited
631
192 Implications for Society
638
Glossary
645
References
659
Author Index
677
Subject Index
681
Copyright

Other editions - View all

Common terms and phrases

About the author (2001)

Rolf Pfeifer is Professor of Computer Science and Director of the Artificial Intelligence Laboratory in the Department of Informatics at the University of Zurich. He is the author of Understanding Intelligence (MIT Press, 1999).

Christian Scheier is a Postdoctoral Fellow at the California Institute of Technology, Pasadena, California.

Bibliographic information