Computing Nonvacuous Generalization Bounds for Deep (Stochastic) Neural Networks with Many More Parameters than Training Data.
Gintare Karolina DziugaiteDaniel M. RoyPublished in: CoRR (2017)
Keyphrases
- training data
- neural network
- generalization bounds
- learning theory
- data dependent
- training process
- data sets
- pattern recognition
- fuzzy logic
- supervised learning
- training set
- maximum likelihood
- machine learning
- statistical learning theory
- training examples
- labeled data
- class labels
- theoretical framework
- neural network model
- unsupervised learning
- model selection
- support vector machine
- classification accuracy
- learning algorithm