D-optimal Bayesian Interrogation for Parameter and Noise Identification of Recurrent Neural Networks
Barnabás PóczosAndrás LörinczPublished in: CoRR (2008)
Keyphrases
- recurrent neural networks
- neural network
- feed forward
- recurrent networks
- complex valued
- artificial neural networks
- neural model
- optimal parameters
- cascade correlation
- reservoir computing
- feedforward neural networks
- nonlinear dynamic systems
- noisy data
- machine learning
- echo state networks
- estimation error
- parameter values
- dynamic programming
- knowledge base
- median filter
- noise level
- single parameter
- noise reduction
- bayesian decision
- optimal solution
- artificial intelligence
- long short term memory