Active Few-Shot Classification: a New Paradigm for Data-Scarce Learning Settings.
Aymane AbdaliVincent GriponLucas DrumetzBartosz BoguslawskiPublished in: CoRR (2022)
Keyphrases
- bayesian methods
- prior knowledge
- data sets
- pattern recognition
- supervised learning
- data analysis
- data collection
- data processing
- machine learning
- incremental learning
- feature selection
- raw data
- learning algorithm
- image data
- data structure
- data quality
- background knowledge
- high dimensional data
- unsupervised learning
- database
- learning process
- decision trees
- video sequences
- input data
- probability distribution
- classification accuracy
- active learning
- classification algorithm
- classification method
- learning tasks
- human experts
- training data
- learned models
- class discrimination