Exploring weight initialization, diversity of solutions, and degradation in recurrent neural networks trained for temporal and decision-making tasks.
Cecilia JarneRodrigo LajePublished in: J. Comput. Neurosci. (2023)
Keyphrases
- training set
- recurrent neural networks
- decision making
- echo state networks
- neural network
- recurrent networks
- feed forward
- decision makers
- spatio temporal
- feedforward neural networks
- complex valued
- artificial neural networks
- cascade correlation
- temporal constraints
- spatial and temporal
- temporal information
- neural model
- hidden layer
- nonlinear dynamic systems
- reservoir computing
- temporal reasoning
- fuzzy logic
- multi layer perceptron
- temporal patterns
- multi objective
- support vector
- video sequences
- optimal solution
- artificial intelligence