Noise Stability is computable and low dimensional.
Anindya DeElchanan MosselJoe NeemanPublished in: CoRR (2017)
Keyphrases
- low dimensional
- high dimensional
- noise level
- manifold learning
- random noise
- high dimensional data
- input space
- dimensionality reduction
- data points
- principal component analysis
- noise model
- noise reduction
- signal to noise ratio
- noisy data
- stability analysis
- arbitrary shape
- multidimensional scaling
- nonlinear dimensionality reduction
- learning algorithm
- median filter
- euclidean space
- dimension reduction
- missing data
- feature space
- image sequences