The Performance of a Matched Subspace Detector That Uses Subspaces Estimated From Finite, Noisy, Training Data.
Nicholas AsendorfRaj Rao NadakuditiPublished in: IEEE Trans. Signal Process. (2013)
Keyphrases
- training data
- linear subspace
- low dimensional
- high dimensional data
- subspace clusters
- lower dimensional
- noisy data
- principal component analysis
- linear projection
- grassmann manifold
- low rank representation
- subspace clustering
- high dimensional
- data sets
- canonical correlations
- feature space
- training set
- decision trees
- hilbert space
- test data
- learning algorithm
- pre trained
- incomplete data
- dimensionality reduction
- input data
- basis vectors
- class labels
- supervised learning
- finite number
- subspace learning
- detection method
- training dataset
- linear combination
- subspace analysis
- feature subspace
- data points
- test set
- training samples
- image space
- label noise
- detection algorithm
- feature extraction
- labeled data
- support vector machine
- finite dimensional
- parameter space
- missing data