Statements (32)
| Predicate | Object |
|---|---|
| gptkbp:instance_of |
gptkb:software_framework
|
| gptkbp:bfsLayer |
5
|
| gptkbp:bfsParent |
gptkb:Language_Models_are_Few-Shot_Learners
gptkb:Vision_Transformers gptkb:Dense_Net-121 |
| gptkbp:application |
Natural language processing
Recommendation systems Image classification |
| gptkbp:benefits |
Reduces the need for labeled data.
|
| gptkbp:challenges |
Domain shift
Semantic gap |
| gptkbp:defines |
A machine learning paradigm that aims to recognize objects or classes that were not seen during training.
|
| https://www.w3.org/2000/01/rdf-schema#label |
Zero-Shot Learning
|
| gptkbp:is_evaluated_by |
Accuracy
F1 Score Precision Recall |
| gptkbp:key |
nan
|
| gptkbp:notable_feature |
Deep learning models
Generative Adversarial Networks (GA Ns) Attribute-based classifiers Semantic output codes Variational Autoencoders (VA Es) |
| gptkbp:philosophy |
Transfer learning
Attribute-based learning |
| gptkbp:related_concept |
gptkb:streaming_service
gptkb:Few-Shot_Learning |
| gptkbp:research_areas |
gptkb:viewpoint
gptkb:Artificial_Intelligence gptkb:software_framework |
| gptkbp:technique |
Generative models
Embedding methods |