Conditional Variational Autoencoder
GPTKB entity
Statements (55)
| Predicate | Object |
|---|---|
| gptkbp:instance_of |
gptkb:television_channel
gptkb:computer |
| gptkbp:bfsLayer |
3
|
| gptkbp:bfsParent |
gptkb:computer
|
| gptkbp:allows |
Mode collapse
Training instability |
| gptkbp:applies_to |
Anomaly detection
Text generation Image generation |
| gptkbp:architectural_style |
Encoder-Decoder
|
| gptkbp:can_be_used_with |
Other generative models
Supervised learning techniques |
| gptkbp:can_create |
Additional information
New data samples |
| gptkbp:developed_by |
gptkb:D._P._Kingma
gptkb:M._Welling |
| gptkbp:has_method |
Generated samples
|
| https://www.w3.org/2000/01/rdf-schema#label |
Conditional Variational Autoencoder
|
| gptkbp:input_output |
Conditional variables
|
| gptkbp:is_different_from |
Standard Autoencoder
|
| gptkbp:is_evaluated_by |
Quantitative metrics
KL divergence Qualitative metrics Reconstruction loss |
| gptkbp:is_explored_in |
gptkb:Workshops
Conferences Research papers Image editing Data augmentation Transfer learning Style transfer Domain adaptation |
| gptkbp:is_implemented_in |
gptkb:Graphics_Processing_Unit
gptkb:Py_Torch |
| gptkbp:is_influenced_by |
Bayesian inference
Neural networks |
| gptkbp:is_part_of |
gptkb:Artificial_Intelligence
Machine learning |
| gptkbp:is_related_to |
gptkb:Generative_Adversarial_Networks
gptkb:computer gptkb:Deep_Learning Latent variable models |
| gptkbp:is_similar_to |
gptkb:computer
|
| gptkbp:is_used_for |
Feature extraction
Semi-supervised learning Data imputation |
| gptkbp:is_used_in |
Natural language processing
Computer vision Recommender systems |
| gptkbp:purpose |
Data generation
Representation learning |
| gptkbp:requires |
Training data
Latent space representation |
| gptkbp:training |
Backpropagation
|
| gptkbp:uses |
Latent variables
|