Cognitive computing systems: Algorithms and applications for networks of neurosynaptic cores
Authors:
- 1 August 2013
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 21614393,p. 1-10
Abstract
Marching along the DARPA SyNAPSE roadmap, IBM unveils a trilogy of innovations towards the TrueNorth cognitive computing system inspired by the brain's function and efficiency. The non-von Neumann nature of the TrueNorth architecture necessitates a novel approach to efficient system design. To this end, we have developed a set of abstractions, algorithms, and applications that are natively efficient for TrueNorth. First, we developed repeatedly-used abstractions that span neural codes (such as binary, rate, population, and time-to-spike), long-range connectivity, and short-range connectivity. Second, we implemented ten algorithms that include convolution networks, spectral content estimators, liquid state machines, restricted Boltzmann machines, hidden Markov models, looming detection, temporal pattern matching, and various classifiers. Third, we demonstrate seven applications that include speaker recognition, music composer recognition, digit recognition, sequence prediction, collision avoidance, optical flow, and eye detection. Our results showcase the parallelism, versatility, rich connectivity, spatio-temporality, and multi-modality of the TrueNorth architecture as well as compositionality of the corelet programming paradigm and the flexibility of the underlying neuron model.Keywords
This publication has 20 references indexed in Scilit:
- Democratic Liquid State Machines for Music Recognition2008
- USARSim: a robot simulator for research and educationProceedings 2001 ICRA. IEEE International Conference on Robotics and Automation (cat. No.01ch37164), 2007
- Reducing the Dimensionality of Data with Neural NetworksScience, 2006
- Collision detection in complex dynamic scenes using an LGMD-based visual neural network with feature enhancementIEEE Transactions on Neural Networks, 2006
- Real-Time Computing Without Stable States: A New Framework for Neural Computation Based on PerturbationsNeural Computation, 2002
- Spiking Neuron Models2002
- CUAVE: A new audio-visual database for multimodal human-computer interface research2002
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998
- Learning representations by back-propagating errorsNature, 1986
- Spatiotemporal energy models for the perception of motionJournal of the Optical Society of America A, 1985