Properties (47)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
software
|
| gptkbp:appearsIn |
image recognition
natural language processing speech recognition |
| gptkbp:canLeadTo |
gradient descent
|
| gptkbp:composedOf |
multiple layers
|
| gptkbp:developedBy |
gptkb:Geoffrey_Hinton
|
| gptkbp:hasCitations |
computationally intensive
requires large datasets |
| gptkbp:hasRelatedPatent |
computer vision
bioinformatics |
| https://www.w3.org/2000/01/rdf-schema#label |
Deep Belief Networks
|
| gptkbp:improves |
feature extraction
|
| gptkbp:includes |
restricted_Boltzmann_machines
|
| gptkbp:isChallengedBy |
overfitting
adversarial attacks vanishing gradients |
| gptkbp:isCounteredBy |
contrastive divergence
|
| gptkbp:isEvaluatedBy |
F1 score
accuracy precision recall |
| gptkbp:isExploredIn |
artificial intelligence
mathematics statistics cognitive science neuroscience |
| gptkbp:isInfluencedBy |
biological neural networks
|
| gptkbp:isInvolvedIn |
gptkb:Keras
Python TensorFlow |
| gptkbp:isPartOf |
deep learning
|
| gptkbp:isPopularIn |
academic research
industry applications |
| gptkbp:isRelatedTo |
neural networks
autoencoders |
| gptkbp:isTestedFor |
backpropagation
layer-wise pre-training |
| gptkbp:isUsedFor |
cross-validation
hierarchical structure data augmentation transfer learning initialize deep networks supervised learning techniques |
| gptkbp:isUsedIn |
recommendation systems
anomaly detection |
| gptkbp:usedFor |
unsupervised learning
|