Negative eigenvalues of the Hessian in deep neural networks.
Guillaume AlainNicolas Le RouxPierre-Antoine ManzagolPublished in: CoRR (2019)
Keyphrases
- neural network
- positive and negative
- pattern recognition
- artificial neural networks
- covariance matrix
- fuzzy systems
- back propagation
- neural network model
- genetic algorithm
- multi layer
- deep learning
- fuzzy logic
- principal component analysis
- neural nets
- data sets
- principal components
- rule extraction
- learning rules
- covariance matrices
- symmetric matrices
- cellular neural networks
- artificial intelligence
- recurrent neural networks
- feed forward
- fault diagnosis
- reinforcement learning
- case study
- information systems