Statements (50)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
Machine Learning Paradigm
|
| gptkbp:application |
gptkb:Natural_Language_Processing
gptkb:Speech_Recognition Medical Diagnosis Image Classification |
| gptkbp:approach |
Data Augmentation
Transfer Learning Meta-Learning Algorithms Metric Learning |
| gptkbp:assesses |
Accuracy on novel classes
Few-shot classification accuracy |
| gptkbp:category |
Supervised Learning
Meta-Learning Semi-Supervised Learning |
| gptkbp:challenge |
Overfitting
Generalization Data Scarcity |
| gptkbp:contrastsWith |
gptkb:Zero-Shot_Learning
One-Shot Learning Traditional Machine Learning |
| gptkbp:enables |
Rapid adaptation to new tasks
|
| gptkbp:field |
gptkb:Machine_Learning
gptkb:artificial_intelligence |
| gptkbp:focusesOn |
Learning with limited labeled data
|
| gptkbp:goal |
Learn from few examples
|
| https://www.w3.org/2000/01/rdf-schema#label |
Few-Shot Learning
|
| gptkbp:key |
Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks (Finn et al., 2017)
Prototypical Networks for Few-shot Learning (Snell et al., 2017) Matching Networks for One Shot Learning (Vinyals et al., 2016) |
| gptkbp:notableFor |
gptkb:Model-Agnostic_Meta-Learning_(MAML)
Prototypical Networks Matching Networks Relation Networks Siamese Networks |
| gptkbp:originatedIn |
2010s
|
| gptkbp:relatedTo |
Transfer Learning
Meta-Learning |
| gptkbp:requires |
Prior Knowledge
Task Similarity |
| gptkbp:trainer |
FewRel
Omniglot miniImageNet tieredImageNet |
| gptkbp:usedIn |
gptkb:robot
Autonomous Vehicles Personalized Recommendation Handwriting Recognition Drug Discovery |
| gptkbp:bfsParent |
gptkb:Large_Language_Models
|
| gptkbp:bfsLayer |
5
|