Agents in the Long Game of AI: Computational Cognitive Modeling for Trustworthy, Hybrid AI
by 作者: Marjorie Mcshane (Author), Sergei Nirenburg (Author), Jesse English (Author)
Publisher Finelybook 出版社: The MIT Press
Publication Date 出版日期: 2024-09-03
Language 语言: English
Pages 页数: 336 pages
ISBN-10 书号: 0262549425
ISBN-13 书号: 9780262549424
Book Description
A novel approach to hybrid AI aimed at developing trustworthy agent collaborators.
The vast majority of current AI relies wholly on machine learning (ML). However, the past thirty years of effort in this paradigm have shown that, despite the many things that ML can achieve, it is not an all-purpose solution to building human-like intelligent systems. One hope for overcoming this limitation is hybrid AI: that is, AI that combines ML with knowledge-based processing. In
Agents in the Long Game of AI, Marjorie McShane, Sergei Nirenburg, and Jesse English present recent advances in hybrid AI with special emphases on content-centric computational cognitive modeling, explainability, and development methodologies.At present, hybridization typically involves sprinkling knowledge into an ML black box. The authors, by contrast, argue that hybridization will be best achieved in the opposite way: by building agents within a cognitive architecture and then integrating judiciously selected ML results. This approach leverages the power of ML without sacrificing the kind of explainability that will foster society’s trust in AI. This book shows how we can develop trustworthy agent collaborators of a type not being addressed by the “ML alone” or “ML sprinkled by knowledge” paradigms—and why it is imperative to do so.
Review
—Shlomo Argamon, Computer Science Department, Illinois Institute of Technology
“The rise of AI based on large language models has led to considerable excitement. But is that approach the best for creating human-like AI systems? McShane and her colleagues argue instead for knowledge-rich cognitive architecture, where language is intertwined with formal representations and reasoning. Unlike large language models, which are trained “at the factory” and cannot easily be updated, their architecture can adapt to people incrementally, a characteristic we expect from human collaborators. This book is an interesting exploration of a possible future direction for AI.”
—Kenneth D Forbus, Walter P. Murphy Professor of Computer Science and Professor of Education, Northwestern University
About the Author
Jesse English is Senior Research Scientist in the Language-Endowed Intelligent Agents Lab at Rensselaer Polytechnic Institute.
