Subclass Deep Neural Networks: Re-enabling Neglected Classes in Deep Network Training for Multimedia Classification.
Nikolaos GkalelisVasileios MezarisPublished in: MMM (1) (2020)
Keyphrases
- neural network
- training process
- multimedia
- radial basis function network
- pattern recognition
- training phase
- training set
- multi layer perceptron
- pattern classification
- network architecture
- class labels
- supervised learning
- recurrent networks
- neural network structure
- deep architectures
- multilayer neural network
- pattern classification problems
- artificial neural networks
- support vector machine svm
- training algorithm
- machine learning
- classification algorithm
- feature extraction
- support vector
- fuzzy logic
- back propagation
- rule extraction
- training samples
- image classification
- network structure
- training data
- backpropagation algorithm
- deep learning
- feed forward neural networks
- learning rules
- classification accuracy
- class imbalance
- complex networks
- neural nets
- feed forward
- feature selection
- training dataset
- multimedia data
- radial basis function
- hopfield network
- training speed
- feature space
- spiking neural networks
- input patterns
- recurrent neural networks
- feature vectors
- pattern languages
- feedforward neural networks
- multimedia content
- multi layer
- training examples