Universal Intelligence: A Definition of Machine Intelligence
E217169
"Universal Intelligence: A Definition of Machine Intelligence" is a foundational paper by Shane Legg (with Marcus Hutter) that formally defines and mathematically characterizes general machine intelligence using concepts from algorithmic information theory and reinforcement learning.
All labels observed (2)
| Label | Occurrences |
|---|---|
| A Formal Measure of Machine Intelligence | 1 |
| Universal Intelligence: A Definition of Machine Intelligence canonical | 1 |
How this entity was disambiguated
This entity first appeared as the object of triple T1922998 — resolving that mention is where its identity was fixed. The disambiguator weighed these candidate entities and picked the highlighted one (or “None”, minting a new entity). This is how homonymy is resolved: the same surface form can point to different entities.
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Universal Intelligence: A Definition of Machine Intelligence Context triple: [Shane Legg, notableWork, Universal Intelligence: A Definition of Machine Intelligence]
-
A.
Superintelligence: Paths, Dangers, Strategies
Superintelligence: Paths, Dangers, Strategies is a 2014 book by philosopher Nick Bostrom that analyzes the potential development of superhuman artificial intelligence and the existential risks and strategic challenges it could pose to humanity.
-
B.
Computing Machinery and Intelligence
"Computing Machinery and Intelligence" is Alan Turing’s landmark 1950 paper that introduced the Turing Test and fundamentally shaped the philosophical and technical foundations of artificial intelligence.
-
C.
"A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence"
"A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence" is the seminal 1955 research proposal by John McCarthy and colleagues that launched the field of artificial intelligence by defining its goals and organizing the landmark 1956 Dartmouth conference.
-
D.
The Age of Intelligent Machines
The Age of Intelligent Machines is a 1990 book by futurist Ray Kurzweil that explores the history, current state, and future implications of artificial intelligence and computing.
-
E.
How to Create a Mind
"How to Create a Mind" is a nonfiction book by futurist Ray Kurzweil that explores the workings of human intelligence and proposes designs for advanced artificial intelligence based on the brain’s principles.
- F. None of above. chosen
- G. Unsure - the case is ambiguous/there is not enough information to decide.
NED2
Entity disambiguation (via description)
gpt-5-mini-2025-08-07
Target entity: Universal Intelligence: A Definition of Machine Intelligence Target entity description: "Universal Intelligence: A Definition of Machine Intelligence" is a foundational paper by Shane Legg (with Marcus Hutter) that formally defines and mathematically characterizes general machine intelligence using concepts from algorithmic information theory and reinforcement learning.
-
A.
Superintelligence: Paths, Dangers, Strategies
Superintelligence: Paths, Dangers, Strategies is a 2014 book by philosopher Nick Bostrom that analyzes the potential development of superhuman artificial intelligence and the existential risks and strategic challenges it could pose to humanity.
-
B.
Computing Machinery and Intelligence
"Computing Machinery and Intelligence" is Alan Turing’s landmark 1950 paper that introduced the Turing Test and fundamentally shaped the philosophical and technical foundations of artificial intelligence.
-
C.
"A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence"
"A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence" is the seminal 1955 research proposal by John McCarthy and colleagues that launched the field of artificial intelligence by defining its goals and organizing the landmark 1956 Dartmouth conference.
-
D.
The Age of Intelligent Machines
The Age of Intelligent Machines is a 1990 book by futurist Ray Kurzweil that explores the history, current state, and future implications of artificial intelligence and computing.
-
E.
How to Create a Mind
"How to Create a Mind" is a nonfiction book by futurist Ray Kurzweil that explores the workings of human intelligence and proposes designs for advanced artificial intelligence based on the brain’s principles.
- F. None of above. chosen
Statements (46)
| Predicate | Object |
|---|---|
| instanceOf |
conference paper
ⓘ
scientific paper ⓘ |
| addresses |
comparison of intelligent agents
ⓘ
evaluation of machine intelligence ⓘ |
| aimsTo |
provide task-independent definition of intelligence
ⓘ
unify different notions of intelligence ⓘ |
| assumes |
bounded reward sequences
ⓘ
computable environments ⓘ |
| author |
Marcus Hutter
ⓘ
Shane Legg ⓘ |
| centralIdea | intelligence as expected reward over all environments weighted by simplicity ⓘ |
| characterizes |
intelligence as ability to achieve goals
ⓘ
intelligence as weighted performance over all computable environments ⓘ |
| citedFor |
formal definition of machine intelligence
ⓘ
foundational work in AGI theory ⓘ |
| contributesTo |
formal foundations of AI
ⓘ
theory of artificial general intelligence ⓘ |
| defines |
formal measure of intelligence
ⓘ
intelligence as performance across environments ⓘ universal intelligence ⓘ |
| field |
algorithmic information theory
ⓘ
artificial intelligence ⓘ machine intelligence ⓘ reinforcement learning ⓘ |
| framework | agent–environment interaction ⓘ |
| hasAbbreviation |
universal intelligence (measure)
ⓘ
linked to:
universal intelligence measure
|
| influencedBy |
Solomonoff induction
ⓘ
algorithmic information theory ⓘ |
| language | English ⓘ |
| proposes | mathematical definition of machine intelligence ⓘ |
| proposesMetric | scalar intelligence measure for agents ⓘ |
| publishedIn |
Proceedings of the International Joint Conference on Artificial Intelligence
ⓘ
linked to:
IJCAI conference
|
| relatedTo |
AIXI model
ⓘ
universal reinforcement learning ⓘ |
| topic |
AIXI-related theory
ⓘ
definition of intelligence ⓘ general machine intelligence ⓘ universal intelligence measure ⓘ |
| usedFor |
benchmarking definitions of intelligence
ⓘ
theoretical analysis of AGI systems ⓘ |
| usesConcept |
Kolmogorov complexity
ⓘ
algorithmic probability ⓘ environment class ⓘ reinforcement learning agent ⓘ reward signal ⓘ universal Turing machine ⓘ
linked to:
Turing machine
|
How these facts were elicited
The pipeline generated the facts above by prompting gpt-5.1 with this entity's name + description and the instruction below.
Instruction
You are a knowledge base construction expert. Given a subject entity and a description of it, return factual statements that you know for the subject as a JSON list of dictionaries(triples), where keys must be "subject", "predicate" and "object". The number of facts may be very high, between 25 to 50 or more, for very popular subjects. For less popular subjects, the number of facts can be very low, like 5 or 10. # Requirements - If you don't know the subject at all, return an empty list. - If the subject is not a named entity, return an empty list. - Include at least one triple where predicate is "instanceOf". - Do not get too wordy. - Separate several objects into multiple triples with one object.
Input
Subject: Universal Intelligence: A Definition of Machine Intelligence Description of subject: "Universal Intelligence: A Definition of Machine Intelligence" is a foundational paper by Shane Legg (with Marcus Hutter) that formally defines and mathematically characterizes general machine intelligence using concepts from algorithmic information theory and reinforcement learning.
Referenced by (2)
Full triples — surface form annotated when it differs from this entity's canonical label.
linked to: Universal Intelligence: A Definition of Machine Intelligence