Timezone: »
Spotlight
An in-silico Neural Model of Dynamic Routing through Neuronal Coherence
Devarajan Sridharan · Brian Percival · john arthur · Kwabena A Boahen
We describe a neurobiologically plausible mechanism to implement dynamic routing using the concept of neuronal communication through neuronal coherence. The model, implemented on a neuromorphic chip, incorporates a three-tier neural network architecture: the lowest tier comprises of raw input representations, the middle tier of the routing neurons, and the topmost tier of invariant output representation. The correct mapping between input and output representations is realized by an appropriate alignment of the phases of their background oscillations by the routing neurons. We demonstrate that our method is able to dramatically reduce the number of connections required from O($N^{3}$) to O($N^{2}$)
Author Information
Devarajan Sridharan (Neurosciences Program, Stanford University)
Brian Percival (Stanford University)
john arthur (Stanford University)
Kwabena A Boahen (Stanford University)
Related Events (a corresponding poster, oral, or spotlight)
-
2007 Poster: An in-silico Neural Model of Dynamic Routing through Neuronal Coherence »
Wed. Dec 5th 06:30 -- 06:40 PM Room
More from the Same Authors
-
2017 Poster: Stochastic and Adversarial Online Learning without Hyperparameters »
Ashok Cutkosky · Kwabena A Boahen -
2016 Poster: Online Convex Optimization with Unconstrained Domains and Losses »
Ashok Cutkosky · Kwabena A Boahen -
2013 Demonstration: Controlling Robot Dynamics With Spiking Neurons »
Samir Menon · Sam Fok · Kwabena A Boahen -
2011 Poster: A Brain-Machine Interface Operating with a Real-Time Spiking Neural Network Control Algorithm »
Julie Dethier · Paul Nuyujukian · Chris Eliasmith · Terrence C Stewart · Shauki A Elasaad · Krishna V Shenoy · Kwabena A Boahen