Modular Neural Simulation with Reconfigurable Hardware
Lead Research Organisation:
Imperial College London
Department Name: Computing
Abstract
This project aims to: (1) provide a development environment for the modular design of complex, large scale neural networks, either for use in computational neuroscience, or for engineering major neural network applications, (2) explore techniques for efficiently and productively mapping such neural networks onto reconfigurable hardware by exploiting parallelism, reconfigurability and design re-use, and (3) demonstrate the resulting hardware and software capabilities by scaling up and accelerating a recently developed brain-inspired control architecture for cognitive robotics, as well as other applications.
Organisations
People |
ORCID iD |
Murray Shanahan (Principal Investigator) | |
Wayne Luk (Co-Investigator) |
Publications
Connor D
(2010)
A computational model of a global neuronal workspace with stochastic connections.
in Neural networks : the official journal of the International Neural Network Society
Fidjeland A
(2009)
NeMo: A Platform for Neural Modelling of Spiking Neurons Using GPUs
Fidjeland A
(2010)
Accelerated simulation of spiking neural networks using GPUs
Fidjeland AK
(2013)
Three tools for the real-time simulation of embodied spiking neural networks using GPUs.
in Neuroinformatics
Gamez D
(2013)
A Neurally Controlled Computer Game Avatar With Humanlike Behavior
in IEEE Transactions on Computational Intelligence and AI in Games
Gamez D
(2012)
iSpike: a spiking neural interface for the iCub robot.
in Bioinspiration & biomimetics
Shanahan M
(2012)
Knotty-centrality: finding the connective core of a complex network.
in PloS one
Wildie M
(2009)
Reconfigurable acceleration of neural models with gap junctions