|
gptkbp:instanceOf
|
recurrent neural network
|
|
gptkbp:activatedBy
|
sign function
|
|
gptkbp:application
|
combinatorial optimization
error correction
optimization problems
pattern recognition
image reconstruction
|
|
gptkbp:canBe
|
continuous
discrete
|
|
gptkbp:capacity
|
approximately 0.15N for N neurons
|
|
gptkbp:convergesTo
|
stable states
|
|
gptkbp:energyFunction
|
gptkb:Lyapunov_function
|
|
gptkbp:energyLandscape
|
multiple minima
|
|
gptkbp:field
|
gptkb:artificial_intelligence
gptkb:machine_learning
computational neuroscience
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
Hopfield Network
|
|
gptkbp:input
|
binary patterns
|
|
gptkbp:introducedIn
|
1982
|
|
gptkbp:learningRule
|
gptkb:Hebbian_learning
|
|
gptkbp:limitation
|
limited storage capacity
spurious states
|
|
gptkbp:namedAfter
|
gptkb:John_Hopfield
|
|
gptkbp:neuronsAre
|
binary
|
|
gptkbp:noSelfConnections
|
true
|
|
gptkbp:output
|
binary patterns
|
|
gptkbp:relatedTo
|
gptkb:Ising_model
Boltzmann machine
autoassociative memory
|
|
gptkbp:roadType
|
fully connected
|
|
gptkbp:stateUpdate
|
asynchronous
synchronous
|
|
gptkbp:usedFor
|
optimization
associative memory
content-addressable memory
|
|
gptkbp:weightMatrix
|
symmetric
|
|
gptkbp:bfsParent
|
gptkb:convolutional_neural_network
|
|
gptkbp:bfsLayer
|
5
|