Guaranteed Convergence of Training Convolutional Neural Networks via Accelerated Gradient Descent.
Shuai ZhangMeng WangSijia LiuPin-Yu ChenJinjun XiongPublished in: CISS (2020)
Keyphrases
- convolutional neural networks
- cost function
- training algorithm
- training process
- convolutional network
- update rule
- feedforward neural networks
- convergence speed
- error function
- data sets
- convergence rate
- training set
- loss function
- online learning
- supervised learning
- training phase
- multi layer perceptron
- probabilistic model
- artificial neural networks
- computer software
- convergence analysis
- feature selection
- neural network