A thesis about theoretical models of super intelligent machines. Includes Hutter's AIXI model, Solomonoff induction, the Universal Intelligence Measure, and the relationship between Goedel incompleteness and artificial intelligence algorithms.
Machine Super Intelligence by Shane Legg
This book develops a formal theory of intelligence, defining it as an agent’s capacity to achieve goals across computable environments and grounding the concept in Kolmogorov complexity, Solomonoff induction and Hutter’s AIXI framework.It shows how these idealised constructs unify prediction, compression and reinforcement learning, yielding a universal intelligence measure while exposing the impracticality of truly optimal agents due to incomputable demands. Finally, it explores how approximate implementations could trigger an intelligence explosion and stresses the profound ethical and existential stakes posed by machines that surpass human capability.
Introduction
Information
- Websitebooks.apple.com
- AuthorsShane Legg
- Published date2011/03/28
Categories
More Items
2023
Harvard EDGE / MLSysBook community, Vijay Janapa Reddi (lead/primary author listed)
MLSysBook (Machine Learning Systems) is an open, community-driven textbook and learning stack for AI systems engineering led by the Harvard EDGE / MLSysBook community. The repository houses the textbook source, TinyTorch (a small educational DL framework), hardware lab kits, and supporting materials to teach how to design, build, benchmark, and deploy real-world machine learning systems.
