Exploiting Low-Rank Tensor-Train Deep Neural Networks Based on Riemannian Gradient Descent With Illustrations of Speech Processing.
Jun QiChao-Han Huck YangPin-Yu ChenJavier TejedorPublished in: CoRR (2022)
Keyphrases
- low rank
- speech processing
- neural network
- trace norm
- high order
- tensor decomposition
- speech recognition
- signal processing
- missing data
- linear combination
- pattern recognition
- multimedia systems
- convex optimization
- matrix completion
- natural language processing
- matrix factorization
- low rank matrix
- kernel matrix
- higher order
- singular value decomposition
- speaker identification
- artificial intelligence
- high dimensional data
- english text
- machine learning
- semi supervised
- image processing
- riemannian manifolds
- pairwise
- manifold learning
- information extraction
- model selection
- language model
- dimensionality reduction
- information retrieval