|
gptkbp:instanceOf
|
gptkb:artificial_intelligence
|
|
gptkbp:activatedBy
|
sigmoid
tanh
|
|
gptkbp:citation
|
Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural computation, 9(8), 1735-1780.
|
|
gptkbp:frameworkSupport
|
gptkb:TensorFlow
gptkb:Keras
gptkb:MXNet
gptkb:Theano
gptkb:PyTorch
|
|
gptkbp:fullName
|
Long Short-Term Memory
|
|
gptkbp:hasApplication
|
musical composition
natural language processing
robotics
bioinformatics
financial forecasting
anomaly detection
video analysis
|
|
gptkbp:hasComponent
|
cell state
forget gate
input gate
output gate
|
|
gptkbp:hasVariant
|
gptkb:Bidirectional_LSTM
gptkb:Convolutional_LSTM
gptkb:Coupled_LSTM
gptkb:Peephole_LSTM
gptkb:Stacked_LSTM
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
LSTM
|
|
gptkbp:introduced
|
gptkb:Sepp_Hochreiter
gptkb:Jürgen_Schmidhuber
|
|
gptkbp:introducedIn
|
1997
|
|
gptkbp:limitation
|
computationally expensive
can overfit on small datasets
difficult to parallelize
|
|
gptkbp:outperformedBy
|
gptkb:Transformer_(in_some_NLP_tasks)
|
|
gptkbp:relatedTo
|
gptkb:GRU
gptkb:transformation
gptkb:RNN
|
|
gptkbp:solvedBy
|
vanishing gradient problem
|
|
gptkbp:type
|
recurrent neural network
|
|
gptkbp:usedFor
|
machine translation
speech recognition
time series prediction
handwriting recognition
sequence modeling
language modeling
|
|
gptkbp:usedIn
|
gptkb:Google_Voice_Search
gptkb:Amazon_Alexa
gptkb:Apple_Siri
|
|
gptkbp:bfsParent
|
gptkb:GraphSAGE
gptkb:Text_Classification
gptkb:Jürgen_Schmidhuber
gptkb:PyTorch_Forecasting
|
|
gptkbp:bfsLayer
|
6
|