Emergence of grid-like representations by training recurrent neural networks to perform spatial localization.
Christopher J. CuevaXue-Xin WeiPublished in: ICLR (Poster) (2018)
Keyphrases
- recurrent neural networks
- recurrent networks
- echo state networks
- feedforward neural networks
- neural network
- spatial information
- feed forward
- reservoir computing
- spatial and temporal
- hidden layer
- artificial neural networks
- nonlinear dynamic systems
- neural model
- grid cells
- complex valued
- feed forward neural networks
- hebbian learning
- training algorithm
- training process
- spatial relations
- spatial data
- supervised learning
- spatio temporal
- cascade correlation
- long short term memory
- grid environment
- nonlinear systems
- semi supervised