Learning Sparse Neural Networks via \ell _0 and T \ell _1 by a Relaxed Variable Splitting Method with Application to Multi-scale Curve Classification.
Fanghui XueJack XinPublished in: WCGO (2019)
Keyphrases
- neural network
- multiscale
- training process
- neural nets
- classification method
- support vector machine svm
- unsupervised learning
- cross validation
- classification accuracy
- learning problems
- pattern recognition
- sparse representation
- training samples
- classification algorithm
- supervised learning
- pattern classification
- active learning
- dictionary learning
- learning algorithm
- roc curve
- multi layer perceptron
- classification process
- machine learning
- feature set
- edge detection
- support vector machine
- prior knowledge
- preprocessing
- objective function
- machine learning algorithms
- energy function
- model selection
- image classification
- sparse coding
- incremental learning
- artificial neural networks
- support vector
- bayesian networks
- markov blanket