Transformed Low-Rank Parameterization Can Help Robust Generalization for Tensor Neural Networks.
Andong WangChao LiMingyuan BaiZhong JinGuoxu ZhouQibin ZhaoPublished in: NeurIPS (2023)
Keyphrases
- low rank
- rank minimization
- low rank matrix
- neural network
- trace norm
- high order
- low rank subspace
- low rank representation
- missing data
- matrix factorization
- robust principal component analysis
- frobenius norm
- tensor decomposition
- convex optimization
- linear combination
- singular value decomposition
- matrix completion
- kernel matrix
- matrix decomposition
- pattern recognition
- high dimensional data
- singular values
- semi supervised
- norm minimization
- higher order
- auxiliary information
- dimensionality reduction
- image segmentation