Training Compact DNNs with ℓ1/2 Regularization.
Anda TangLingfeng NiuJianyu MiaoPeng ZhangPublished in: Pattern Recognit. (2023)
Keyphrases
- early stopping
- online learning
- information systems
- training algorithm
- neural network
- supervised learning
- case study
- virtual environment
- parameter selection
- stochastic gradient descent
- training set
- multi class
- maximum likelihood
- training examples
- test set
- training process
- decision trees
- energy functional
- structured prediction