Statements (52)
| Predicate | Object |
|---|---|
| gptkbp:instance_of |
gptkb:machine_learning
|
| gptkbp:application |
Natural language processing
Speech recognition Image classification |
| gptkbp:benefits |
Reduced training time
Generalization to new tasks Limited interpretability Lower data requirements Requires careful tuning |
| gptkbp:challenges |
Overfitting
Data scarcity |
| gptkbp:defines |
A type of machine learning where the model is trained on a small number of examples.
|
| gptkbp:evaluates |
Accuracy
Precision Recall Cross-validation F1 score Holdout validation Leave-one-out validation |
| gptkbp:field |
gptkb:neural_networks
gptkb:Deep_Learning Statistical Learning |
| gptkbp:future_prospects |
Expansion to more complex tasks
Improvement in model architectures Integration with reinforcement learning |
| gptkbp:goal |
Enhance model robustness
Facilitate learning from limited data Improve learning efficiency |
| https://www.w3.org/2000/01/rdf-schema#label |
Few-Shot Learning
|
| gptkbp:is_a_tool_for |
gptkb:Tensor_Flow
gptkb:Keras gptkb:Py_Torch |
| gptkbp:is_compared_to |
Zero-shot learning
One-shot learning Traditional machine learning |
| gptkbp:philosophy |
gptkb:Matching_networks
gptkb:Siamese_networks Prototypical networks |
| gptkbp:provides_information_on |
gptkb:CUB-200-2011
gptkb:Mini_Image_Net gptkb:Omniglot Fungi dataset |
| gptkbp:research_areas |
gptkb:Computer_Vision
gptkb:Artificial_Intelligence gptkb:robotics |
| gptkbp:technique |
Transfer learning
Meta-learning |
| gptkbp:theory |
Bayesian inference
Graphical models Variational inference |
| gptkbp:bfsParent |
gptkb:GPT-3
|
| gptkbp:bfsLayer |
5
|