Implicit Self-Regularization in Deep Neural Networks: Evidence from Random Matrix Theory and Implications for Learning.
Charles H. MartinMichael W. MahoneyPublished in: CoRR (2018)
Keyphrases
- neural network
- learning algorithm
- learning process
- inductive inference
- learning tasks
- online learning
- random matrix theory
- supervised learning
- active learning
- genetic algorithm
- reinforcement learning
- training data
- learning systems
- image processing
- neural nets
- multi task
- e learning
- learning rules
- deep architectures
- connectionist systems