Autonomous learning algorithm for fully connected recurrent networks.
Edouard LeclercqFabrice DruauxDimitri LefebvrePublished in: ESANN (2003)
Keyphrases
- fully connected
- recurrent networks
- learning algorithm
- recurrent neural networks
- feed forward
- back propagation
- conditional random fields
- activation function
- neural network
- biologically inspired
- scale free
- active learning
- machine learning algorithms
- machine learning
- training data
- learning rate
- reinforcement learning
- training examples
- multi modal
- maximum likelihood
- graphical models
- higher order
- decision making