An AGI Brain for a Robot provides a working model of a robot brain, the first book to give a detailed account and practical demonstration of an Artificial General Intelligence (AGI). Readers will learn how to design and implement a brain in fast parallel hardware and embody it in the head of a robot moving in the real world. Associative learning is shown to be a powerful technique for novelty seeking, language learning, and planning. This book is for neuroscientists, robot designers, psychologists, philosophers and anyone curious about the evolution of the human brain and its specialized functions.
The overarching message of this book is that an AGI, as the brain of a robot, is within our grasp and would work like our own brains. The featured brain, called PP, is not a computer program. Instead, PP is a collection of networks of associations built from J. A. Fodor's modules and the author's Groups. The associations are acquired by intimate interaction between PP in its robot body and the real world. Simulations of PP in one of two robots in a simple world demonstrate PP learning from the second robot, which is under human control. PP has free will from its novelty goals, learns to use language, employs working memory, makes plans, and follows them. It is argued that the ability to answer questions about what it is doing, what it feels, and what it intends to do will eventually reveal conscious awareness.
- Explains how to design and implement a robot brain with real world functions
- Describes fast parallel hardware for an AGI brain
- Exhibits free will, language learning, working memory, and planning
- Introduces multiple context associative learning using modules and Groups
- Provides a Java computer program and output data files via a companion website
1. Brain, Body and World 2. Groups 3. Interacting with PP 4. Learning to Take Turns 5. No Approval from Teacher 6. Experiments from the Past 7. Consciousness
Appendix 1. Squashes in First Interaction 2. Touching 3. Excerpts from the Interactions 4. PP program and Architecture
Dr. Andreae obtained his PhD from Imperial College, London University in 1955. His research in Artificial Intelligence began at Standard Telecommunication Laboratories, Harlow, Essex, and he published a paper on his first learning machine, STeLLA, in 1963. In 1966 he moved from England to the Department of Electrical Engineering, University of Canterbury in Christchurch, NZ, where he continued his research and invented his second learning machine, PurrPuss, later shortened to PP. He has continued the research during retirement. Dr. Andreae has previously written two books on his research: Thinking with the Teachable Machine (Academic Press, 1977) and Associative Learning for a Robot Intelligence (Imperial College Press, 1998).