Skip to main content

2012 | Buch

Theoretical Foundations of Artificial General Intelligence

insite
SUCHEN

Über dieses Buch

This book is a collection of writings by active researchers in the field of Artificial General Intelligence, on topics of central importance in the field. Each chapter focuses on one theoretical problem, proposes a novel solution, and is written in sufficiently non-technical language to be understandable by advanced undergraduates or scientists in allied fields. This book is the very first collection in the field of Artificial General Intelligence (AGI) focusing on theoretical, conceptual, and philosophical issues in the creation of thinking machines. All the authors are researchers actively developing AGI projects, thus distinguishing the book from much of the theoretical cognitive science and AI literature, which is generally quite divorced from practical AGI system building issues. And the discussions are presented in a way that makes the problems and proposed solutions understandable to a wide readership of non-specialists, providing a distinction from the journal and conference-proceedings literature. The book will benefit AGI researchers and students by giving them a solid orientation in the conceptual foundations of the field (which is not currently available anywhere); and it would benefit researchers in allied fields by giving them a high-level view of the current state of thinking in the AGI field. Furthermore, by addressing key topics in the field in a coherent way, the collection as a whole may play an important role in guiding future research in both theoretical and practical AGI, and in linking AGI research with work in allied disciplines

Inhaltsverzeichnis

Frontmatter
Chapter 1. Introduction: What Is the Matter Here?
Abstract
This chapter provides a general introduction to the volume, giving an overview of the AGI field and the current need for exploration and clarification of its foundations, and briefly summarizing the contents of the various chapters.
Pei Wang, Ben Goertzel
Chapter 2. Artificial Intelligence and Cognitive Modeling Have the Same Problem
Abstract
Cognitive modelers attempting to explain human intelligence share a puzzle with artificial intelligence researchers aiming to create computers that exhibit human-level intelligence: how can a system composed of relatively unintelligent parts (such as neurons or transistors) behave intelligently? I argue that although cognitive science has made significant progress towards many of its goals, that solving the puzzle of intelligence requires special standards and methods in addition to those already employed in cognitive science. To promote such research, I suggest creating a subfield within cognitive science called intelligence science and propose some guidelines for research addressing the intelligence puzzle.
Nicholas L. Cassimatis
Chapter 3. Psychometric Artificial General Intelligence: The Piaget-MacGuyver Room
Abstract
Psychometric AGI (PAGI) is the brand of AGI that anchors AGI science and engineering to explicit tests, by insisting that for an information-processing (i-p) artifact to be rationally judged generally intelligent, creative, wise, and so on, it must pass a suitable, well-defined test of such mental power(s). Under the tent of PAGI, and inspired by prior thinkers, we introduce the Piaget-MacGyver Room (PMR), which is such that, an i-p artifact can credibly be classified as general-intelligent if and only if it can succeed on any test constructed from the ingredients in this room.
Selmer Bringsjord, John Licato
Chapter 4. Beyond the Octopus: From General Intelligence Toward a Human-Like Mind
Abstract
General intelligence varies with species and environment. Octopuses are highly intelligent, sensing and rapidly learning the complex properties of their world. But as asocial creatures, all their learned knowledge dies with them. Humans, on the other hand, are exceedingly social, gathering much more complex information and sharing it with others in their family, community and wider culture. In between those extremes there are several distinct types, or levels, of reasoning and information sharing that we characterize as a metaphorical “ladder” of intelligence.
Sam S. Adams, Steve Burbeck
Chapter 5. One Decade of Universal Artificial Intelligence
Abstract
The first decade of this century has seen the nascency of the first mathematical theory of general artificial intelligence. This theory of Universal Artificial Intelligence (UAI) has made significant contributions to many theoretical, philosophical, and practical AI questions. In a series of papers culminating in book an exciting sound and complete mathematical model for a super intelligent agent (AIXI) has been developed and rigorously analyzed.
Marcus Hutter
Chapter 6. Deep Reinforcement Learning as Foundation for Artificial General Intelligence
Abstract
Deep machine learning and reinforcement learning are two complementing fields within the study of intelligent systems. When combined, it is argued that they offer a promising path for achieving artificial general intelligence (AGI). This chapter outlines the concepts facilitating such merger of technologies and motivates a framework for building scalable intelligent machines. The prospect of utilizing custom neuromorphic devices to realize large-scale deep learning architectures is discussed, paving the way for achieving human level AGI.
Itamar Arel
Chapter 7. The LIDA Model as a Foundational Architecture for AGI
Abstract
Artificial intelligence (AI) initially aimed at creating “thinking machines,” that is, computer systems having human level general intelligence. However, AI research has until recently focused on creating intelligent, but highly domain-specific, systems. Currently, researchers are again undertaking the original challenge of creating AI systems (agents) capable of human-level intelligence, or “artificial general intelligence” (AGI).
Usef Faghihi, Stan Franklin
Chapter 8. The Architecture of Human-Like General Intelligence
Abstract
By exploring the relationships between different AGI architectures, one can work toward a holistic cognitive model of human-level intelligence. In this vein, here an integrative architecture diagram for human-like general intelligence is proposed, via merging of lightly modified version of prior diagrams including Aaron Sloman’s high-level cognitive model, Stan Franklin and the LIDA group’s model of working memory and the cognitive cycle, Joscha Bach and Dietrich Dörner’s Psi model of motivated action and cognition, James Albus’s three-hierarchy intelligent robotics model, and the author’s prior work on cognitive synergy in deliberative thought and metacognition, along with ideas from deep learning and computational linguistics.
Ben Goertzel, Matt Iklé, Jared Wigmore
Chapter 9. A New Constructivist AI: From Manual Methods to Self-Constructive Systems
Abstract
The development of artificial intelligence (AI) systems has to date been largely one of manual labor. This constructionist approach to AI has resulted in systems with limited-domain application and severe performance brittleness. No AI architecture to date incorporates, in a single system, the many features that make natural intelligence general-purpose, including system-wide attention, analogy-making, system-wide learning, and various other complex transversal functions.
Kristinn R. Thórisson
Chapter 10. Towards an Actual Gödel Machine Implementation: a Lesson in Self-Reflective Systems
Abstract
Recently, interest has been revived in self-reflective systems in the context of Artificial General Intelligence (AGI). An AGI system should be intelligent enough to be able to reason about its own program code, and make modifications where it sees fit, improving on the initial code written by human programmers. A pertinent example is the Gödel Machine, which employs a proof searcher—in parallel to its regular problem solves duties—to find a self-rewrite of which it can prove that it will be beneficial.
Bas R. Steunebrink, Jürgen Schmidhuber
Chapter 11. Artificial General Intelligence Begins With Recognition: Evaluating the Flexibility of Recognition
Abstract
Many types of supervised recognition algorithms have been developed over the past half century. However, it remains difficult to compare their flexibility and ability to reason. Part of the difficulty is the need of a good definition of flexibility. This chapter is dedicated to defining and evaluating flexibility in recognition.
Tsvi Achler
Chapter 12. Theory Blending as a Framework for Creativity in Systems for General Intelligence
Abstract
Being creative is a central property of humans in solving problems, adapting to new states of affairs, applying successful strategies in previously unseen situations, or coming up with new conceptualizations. General intelligent systems should have the potential to realize such forms of creativity to a certain extent. We think that creativity and productivity issues can be best addressed by taking cognitive mechanisms into account, such as analogy making, concept blending, computing generalizations and the like.
Maricarmen Martinez, Tarek R. Besold, Ahmed Abdel-Fattah, Helmar Gust, Martin Schmidt, Ulf Krumnack, Kai-Uwe Kühnberger
Chapter 13. Modeling Motivation and the Emergence of Affect in a Cognitive Agent
Abstract
By exploring the relationships between different AGI architectures, one can work toward a holistic cognitive model of human-level intelligence. In this vein, here an integrative architecture diagram for human-like general intelligence is proposed, via merging of lightly modified version of prior diagrams including Aaron Sloman’s high-level cognitive model, Stan Franklin and the LIDA group’s model of working memory and the cognitive cycle, Joscha Bach and Dietrich Dörner’s Psi model of motivated action and cognition, James Albus’s three-hierarchy intelligent robotics model, and the author’s prior work on cognitive synergy in deliberative thought and metacognition, along with ideas from deep learning and computational linguistics.
Joscha Bach
Chapter 14. AGI and Machine Consciousness
Abstract
This review discusses some of main issues to be addressed to design a conscious AGI agent: the agent’s sense of the body, the interaction with the environment, the agent’s sense of time, the free will of the agent, the capability for the agent to have some form of experience, and finally the relationship between consciousness and creativity.
Antonio Chella, Riccardo Manzotti
Chapter 15. Human and Machine Consciousness as a Boundary Effect in the Concept Analysis Mechanism
Abstract
To solve the hard problem of consciousness we observe that any cognitive system of sufficient power must get into difficulty when it tries to analyze consciousness concepts, because the mechanism that does the analysis will “bottom out” in such a way as to make the system declare these concepts to be both real and ineffable. Rather than use this observation to dismiss consciousness as an artifact, we propose a unifying interpretation that allows consciousness to be explicable at a meta level, while at the same time being mysterious and inexplicable on its own terms.
Richard Loosemore
Chapter 16. Theories of Artificial Intelligence—Meta-Theoretical considerations
Abstract
This chapter addresses several central meta-theoretical issues of AI and AGI. After analyzing the nature of the field, three criteria for desired theories are proposed: correctness, concreteness, and compactness. The criteria are clarified in the AI context, and using them, the current situation in the field is evaluated.
Pei Wang
Backmatter
Metadaten
Titel
Theoretical Foundations of Artificial General Intelligence
herausgegeben von
Pei Wang
Ben Goertzel
Copyright-Jahr
2012
Verlag
Atlantis Press
Electronic ISBN
978-94-91216-62-6
Print ISBN
978-94-91216-61-9
DOI
https://doi.org/10.2991/978-94-91216-62-6

Premium Partner