Classifier Probes May Just Learn from Linear Context Features.
Jenny KunzMarco KuhlmannPublished in: COLING (2020)
Keyphrases
- feature set
- feature space
- classification process
- svm classifier
- classification models
- feature extractors
- feature values
- extracted features
- feature vectors
- maximum entropy classifier
- linear svm
- classification method
- feature extraction
- image features
- feature reduction
- decision trees
- weak learners
- input features
- neural network
- machine learning
- feature extractor
- supervised classifiers
- multiple classifiers
- training examples
- feature weighting
- training data
- linear classifiers
- low level
- co occurrence
- contextual information