Finite-State Reber Automaton and the Recurrent Neural Networks Trained in Supervised and Unsupervised Manner.
Michal CernanskýLubica BenuskováPublished in: ICANN (2001)
Keyphrases
- finite state
- unsupervised manner
- recurrent neural networks
- echo state networks
- unsupervised learning
- tree automata
- markov chain
- markov decision processes
- neural network
- reservoir computing
- recurrent networks
- feed forward
- model checking
- optimal policy
- semi supervised
- supervised learning
- artificial neural networks
- training set
- learning algorithm
- multilayer perceptron
- cascade correlation
- multi layer perceptron
- nonlinear dynamic systems
- decision trees
- machine learning
- expectation maximization
- feature selection
- context free
- finite automata
- image segmentation
- reinforcement learning
- search algorithm