Embroid: Unsupervised Prediction Smoothing Can Improve Few-Shot Classification.
Neel GuhaMayee F. ChenKush BhatiaAzalia MirhoseiniFrederic SalaChristopher RéPublished in: CoRR (2023)
Keyphrases
- supervised learning
- unsupervised learning
- supervised classification
- classification scheme
- machine learning
- prediction accuracy
- visual features
- unsupervised clustering
- image classification
- decision trees
- text classification
- support vector machine
- feature vectors
- feature space
- improve the prediction accuracy
- support vector machine svm
- pattern recognition
- class labels
- classification rules
- nearest neighbour algorithm
- neural networks and support vector machines
- partially supervised
- classification algorithm
- training samples
- prediction error
- automatic classification
- support vector
- feature extraction
- multi layer perceptron
- data sets
- training set
- semi supervised
- classification accuracy