Occlusion Sensitivity Analysis with Augmentation Subspace Perturbation in Deep Feature Space.
Pedro ValoisKoichiro NiinumaKazuhiro FukuiPublished in: CoRR (2023)
Keyphrases
- sensitivity analysis
- feature space
- low dimensional
- high dimensional
- kernel based nonlinear
- managerial insights
- principal component analysis
- lower dimensional
- dimensionality reduction
- high dimensionality
- input space
- classification accuracy
- input data
- kernel function
- mean shift
- high dimensional data
- dimension reduction
- dot product
- kernel methods
- feature set
- feature vectors
- feature selection
- influence diagrams
- data points
- canonical correlations
- support vector machine
- variational inequalities
- training set
- subspace learning
- feature extraction
- image representation
- high dimensional feature space