Less is More: Dimension Reduction Finds On-Manifold Adversarial Examples in Hard-Label Attacks.
Washington GarciaPin-Yu ChenHamilton Scott ClouseSomesh JhaKevin R. B. ButlerPublished in: SaTML (2023)
Keyphrases
- dimension reduction
- manifold learning
- low dimensional
- high dimensional
- manifold embedding
- feature space
- feature extraction
- principal component analysis
- high dimensional data
- random projections
- high dimensional problems
- dimensionality reduction
- intrinsic dimension
- nonlinear manifold
- generative topographic mapping
- singular value decomposition
- manifold learning algorithm
- high dimensionality
- feature selection
- discriminative information
- linear discriminant analysis
- unsupervised learning
- pattern recognition
- head pose estimation
- cluster analysis
- dimension reduction methods
- high dimensional data analysis
- metric learning
- locally linear embedding
- feature vectors
- preprocessing