ProxSGD: Training Structured Neural Networks under Regularization and Constraints.
Yang YangYaxiong YuanAvraam ChatzimichailidisRuud J. G. van SlounLei LeiSymeon ChatzinotasPublished in: ICLR (2020)
Keyphrases
- neural network
- training process
- feedforward neural networks
- feed forward neural networks
- multi layer perceptron
- training algorithm
- back propagation
- artificial neural networks
- pattern recognition
- early stopping
- neural network training
- backpropagation algorithm
- constraint satisfaction
- multi layer
- training phase
- constrained optimization
- recurrent neural networks
- neural nets
- feed forward
- constraint programming
- structured data
- online learning
- error back propagation
- global constraints
- training examples
- potential functions
- linear programming
- recurrent networks
- fuzzy logic
- active learning
- neural network structure
- feature space