2014 · 41 citations · 7 references
EngineeringComputer EngineeringComputer SystemsField-programmable Gate ArraysNeural EngineeringReconfigurable ComputingFpga ArchitectureHardware AccelerationNeural Circuit ModelingNeural CircuitsSpiking Neural NetworksConductanceConductance-based ModelsConductance-based Neuron ModelsHardware ResourcesSpiking Network ArchitecturesSpiking Neural Network ModelsSpiking Neuron ModelsThroughput
We present an FPGA design framework for large-scale spiking neural networks, particularly the ones with a high-density of connections or all-to-all connections. The proposed FPGA design framework is based on a reconfigurable neural layer, which is implemented using a time-multiplexing approach to achieve up to 200,000 virtual neurons with one physical neuron using only a fraction of the hardware resources in commercial-off-the-shelf FPGAs (even entry level ones). Rather than using a mathematical computational model, the physical neuron was efficiently implemented with a conductance-based model, of which the parameters were randomised between neurons to emulate the variance in biological neurons. Besides these building blocks, the proposed time-multiplexed reconfigurable neural layer has an address buffer, which will generate a fixed random weight for each connection on the fly for incoming spikes. This structure effectively reduces the usage of memory. After presenting the architecture of the proposed neural layer, we present a network with 23 proposed neural layers, each containing 64k neurons, yielding 1.5 M neurons and 92 G synapses with a total spike throughput of 1.2T spikes/s, while running in real-time on a Virtex 6 FPGA.
7
Simple model of spiking neurons
Eugene M. Izhikevich · IEEE Transactions on Neural Networks · 2003 · 4.8K citations
Challenges for large-scale implementations of spiking neural networks on FPGAs
Liam Maguire, T.M. McGinnity, Brendan Glackin et al. · Neurocomputing · 2007 · 179 citations