Statements (23)
Predicate | Object |
---|---|
gptkbp:instanceOf |
gptkb:model
theoretical agent |
gptkbp:actionSelection |
maximizes expected future reward
|
gptkbp:approximation |
gptkb:AIXItl
|
gptkbp:basedOn |
gptkb:reinforcement_learning
gptkb:Solomonoff_induction |
gptkbp:computability |
incomputable
|
gptkbp:describedBy |
gptkb:Universal_Artificial_Intelligence:_Sequential_Decisions_based_on_Algorithmic_Probability
|
gptkbp:environmentModel |
unknown environment
|
gptkbp:field |
gptkb:artificial_intelligence
gptkb:machine_learning theoretical computer science |
gptkbp:firstDescribed |
2000
|
gptkbp:goal |
maximize expected reward
|
https://www.w3.org/2000/01/rdf-schema#label |
AIXI agent
|
gptkbp:limitation |
not computable in practice
|
gptkbp:proposedBy |
gptkb:Marcus_Hutter
|
gptkbp:relatedTo |
gptkb:AGI
universal intelligence |
gptkbp:rewardSignal |
external environment
|
gptkbp:uses |
algorithmic probability
|
gptkbp:bfsParent |
gptkb:Universal_Artificial_Intelligence
|
gptkbp:bfsLayer |
7
|