Intelligence via ultrafilters: structural properties of some intelligence comparators of deterministic Legg-Hutter agents
Journal of Artificial General Intelligence 10 (1):24-45 (2019)
Abstract
Legg and Hutter, as well as subsequent authors, considered intelligent agents through the lens of interaction with reward-giving environments, attempting to assign numeric intelligence measures to such agents, with the guiding principle that a more intelligent agent should gain higher rewards from environments in some aggregate sense. In this paper, we consider a related question: rather than measure numeric intelligence of one Legg- Hutter agent, how can we compare the relative intelligence of two Legg-Hutter agents? We propose an elegant answer based on the following insight: we can view Legg-Hutter agents as candidates in an election, whose voters are environments, letting each environment vote (via its rewards) which agent (if either) is more intelligent. This leads to an abstract family of comparators simple enough that we can prove some structural theorems about them. It is an open question whether these structural theorems apply to more practical intelligence measures.Author's Profile
My notes
Similar books and articles
Universal intelligence: A definition of machine intelligence.Shane Legg & Marcus Hutter - 2007 - Minds and Machines 17 (4):391-444.
One decade of universal artificial intelligence.Marcus Hutter - 2012 - In Pei Wang & Ben Goertzel (eds.), Theoretical Foundations of Artificial General Intelligence. Springer. pp. 67--88.
The Superintelligent Will: Motivation and Instrumental Rationality in Advanced Artificial Agents. [REVIEW]Nick Bostrom - 2012 - Minds and Machines 22 (2):71-85.
A Value-Sensitive Design Approach to Intelligent Agents.Steven Umbrello & Angelo Frank De Bellis - 2018 - In Roman Yampolskiy (ed.), Artificial Intelligence Safety and Security. New York, NY, USA: CRC Press. pp. 395-410.
"Every man in his notions" or alchemists' discussion on artificial intelligence.Mariusz Flasiński - 1997 - Foundations of Science 2 (1):107-121.
Safety Engineering for Artificial General Intelligence.Roman Yampolskiy & Joshua Fox - 2013 - Topoi 32 (2):217-226.
Consciousness, intentionality, and intelligence: Some foundational issues for artificial intelligence.Murat Aydede & Guven Guzeldere - 2000 - Journal of Experimental and Theoretical Artificial Intelligence 12 (3):263-277.
Re-creating the Philosopher’s Mind: Artificial Life from Artificial Intelligence.Maurice H. T. Ling - 2012 - Human-Level Intelligence 2:1.
The "Artificial Life" Route to "Artificial Intelligence": Building Situated Embodied Agents.Luc Steels & Rodney Brooks (eds.) - 1995 - Hillsdale, NJ: Lawrence Erlbaum Associates.
Proceedings of the Second Conference on Artificial General Intelligence.B. Goertzel, P. Hitzler & M. Hutter (eds.) - 2009 - Atlantis Press.
Analytics
Added to PP
2019-10-22
Downloads
1,182 (#5,689)
6 months
39 (#34,313)
2019-10-22
Downloads
1,182 (#5,689)
6 months
39 (#34,313)
Historical graph of downloads
Author's Profile
Citations of this work
The Archimedean trap: Why traditional reinforcement learning will probably not yield AGI.Samuel Allen Alexander - 2020 - Journal of Artificial General Intelligence 11 (1):70-85.
Measuring Intelligence and Growth Rate: Variations on Hibbard's Intelligence Measure.Samuel Alexander & Bill Hibbard - 2021 - Journal of Artificial General Intelligence 12 (1):1-25.
References found in this work
Universal intelligence: A definition of machine intelligence.Shane Legg & Marcus Hutter - 2007 - Minds and Machines 17 (4):391-444.
An impossibility theorem for amalgamating evidence.Jacob Stegenga - 2013 - Synthese 190 (12):2391-2411.
Measuring universal intelligence: Towards an anytime intelligence test.José Hernández-Orallo & David L. Dowe - 2010 - Artificial Intelligence 174 (18):1508-1539.