Negative Margin Matters: Understanding Margin in Few-Shot Classification.
Bin LiuYue CaoYutong LinQi LiZheng ZhangMingsheng LongHan HuPublished in: ECCV (4) (2020)
Keyphrases
- support vector
- decision boundary
- training set
- automatic classification
- pattern classification
- soft margin
- classification accuracy
- support vector machine
- maximum margin
- supervised learning
- feature selection
- text classification
- classification systems
- feature space
- classification scheme
- video sequences
- machine learning
- support vector machine svm
- training data
- decision trees
- generalization ability
- classification method
- classification algorithm
- class labels
- classification process
- pattern recognition
- image classification
- preprocessing
- feature vectors
- margin maximization
- features extraction
- standard svm
- detection algorithm
- visual features
- model selection
- data points
- feature extraction