Over-complete representations on recurrent neural networks can support persistent percepts.
Shaul DruckmannDmitri B. ChklovskiiPublished in: NIPS (2010)
Keyphrases
- recurrent neural networks
- feed forward
- neural network
- reservoir computing
- artificial neural networks
- end users
- echo state networks
- complex valued
- multiple representations
- recurrent networks
- higher level
- cascade correlation
- nonlinear dynamic systems
- long short term memory
- real time
- hebbian learning
- feedforward neural networks
- computational model
- e learning
- artificial intelligence