MetAdapt: Meta-Learned Task-Adaptive Architecture for Few-Shot Classification.
Sivan DovehEli SchwartzChao XueRogério Schmidt FerisAlexander M. BronsteinRaja GiryesLeonid KarlinskyPublished in: CoRR (2019)
Keyphrases
- classification accuracy
- pattern classification
- decision trees
- pattern recognition
- feature vectors
- machine learning
- classification method
- classification algorithm
- automatic classification
- classification scheme
- management system
- support vector machine
- text classification
- learning phase
- support vector machine svm
- support vector
- feature space
- unsupervised learning
- video data
- machine learning algorithms
- classification models
- class labels
- meta level
- classification process
- classification systems
- neural network
- shot change detection
- change detection
- benchmark datasets
- training samples
- model selection
- supervised learning
- semi supervised
- training set
- preprocessing
- feature extraction