Loihi: A Neuromorphic Manycore Processor with On-Chip Learning
Citations
877 citations
697 citations
Cites background or methods from "Loihi: A Neuromorphic Manycore Proc..."
...of the hardware platform, (iii) 22 Table 4 Comparison between selected neuromorphic processors, ordered by neuron model type. Processor SpiNNaker TrueNorth Loihi DYNAP Braindrop Reference [258] [259] [260] [261] [262] Manufacturer U. Manchester IBM Intel aiCTX Stanford U. Year 2011 2014 2018 2017 2018 Neuron model Software Digital Digital Analog Analog On-chip learning Yes No Yes No No CMOS technol. 13...
[...]
...g. Each chip simulates up to 131 thousand spiking neurons and 130M synapses. A learning engine in each neuromorphic core updates each synapse using rules that includes STDP and reinforcement learning [260]. Nonspiking networks can be trained in TensorFlow and approximated by spiking networks for Loihi using the Nengo Deep Learning toolkit from Applied Brain Research [271]. There are plans to build a la...
[...]
614 citations
568 citations
545 citations
References
11,413 citations
"Loihi: A Neuromorphic Manycore Proc..." refers background in this paper
...On postsynaptic spike (3) where F(x) is some approximation of e-x/τ ⋅ H(x), for constants A− < 0, A+ > 0, and τ > 0....
[...]
...Population-mapped neurons produce population spike messages whose axon_id fields identify (1) the destination population Pdst, (2) the source neuron index i ∈ Psrc within the source population, and (3) the particular edge connecting between Tsrc and Tdst when there is more than one....
[...]
3,253 citations
1,171 citations
607 citations
526 citations
Additional excerpts
...P z z S V C = = = + + ∏ (4) where z is the transformed synaptic variable (either wgt, dly, or tag), Vi,j refers to some choice of input variable available to the learning engine, and Ci,j and Si are microcode-specified signed constants....
[...]