Project ORGANIC: ORGANIC: Self-organized recurrent neural learing for language processing
2009-04-01 – 2012-03-31
- Abstract
ORGANICE adopts principles of cortical architectures and self-organizing neurodynamics for the design of a new type of cognitive architectures. The principle inovations are deep multilevel learning based on recurrent networks using purely dynamic representations and trained both supervised and unsupervised. These techniques will be applied to large vocabulary speech recognition in noisy environments, end on handwriting recognition.
Show
Sort by
-
Continuous digit recognition in noise: reservoirs can do an excellent job!
-
Learning slow features with reservoir computing for biologically-inspired robot localization
-
Connected digit recognition by means of Reservoir Computing
-
Phoneme recognition with large hierarchical reservoirs
-
Memory in linear recurrent neural networks in continuous time
-
Fast and exact simulation methods applied on a broad range of neuron models