Mapping Spiking Neural Networks onto a Manycore Neuromorphic Architecture
We present a compiler for Loihi, a novel manycore neuromorphic processor that features a programmable, on-chip learning engine for training and executing spiking neural networks (SNNs). An SNN is distinguished from other neural networks in that (1) its independent computing units, or "neurons", communicate with others only through spike messages; and (2) each neuron evaluates local learning rules, which are functions of spike arrival and departure timings, to modify its local state. The collective neuronal state dynamics of an SNN form a nonlinear dynamical system that can be cast as an unconventional model of computation. To realize such an SNN on Loihi requires each constituent neuron to locally store and independently update its own spike timing information. However, each Loihi core has limited resources for this purpose and these must be shared by neurons assigned to the same core. In this work, we present a compiler for Loihi that maps the neurons of an SNN onto and across Loihi's cores efficiently. We show that a poor neuron-to-core mapping can incur significant energy costs and address this with a greedy algorithm that compiles SNNs onto Loihi in a power-efficient manner. In so doing, we highlight the need for further development of compilers for this new, emerging class of architectures.
Wed 20 Jun Times are displayed in time zone: Eastern Time (US & Canada) change
11:00 - 12:15: Emerging HardwarePLDI Research Papers at Grand Ballroom CD Chair(s): Ryan R. NewtonIndiana University | |||
11:00 - 11:25 Talk | Persistency for Synchronization-Free Regions PLDI Research Papers Vaibhav GogteUniversity of Michigan, USA, Stephan DiestelhorstARM Research, UK, William WangArm Research, UK, Satish NarayanasamyUniversity of Michigan, Peter M. ChenUniversity of Michigan, USA, Thomas F. WenischUniversity of Michigan, USA Media Attached | ||
11:25 - 11:50 Talk | Write-Rationing Garbage Collection for Hybrid Memories PLDI Research Papers Shoaib AkramGhent University, Jennifer B. SartorVrije Universiteit Brussel, Kathryn S McKinleyGoogle, Lieven EeckhoutGhent University, Belgium Media Attached | ||
11:50 - 12:15 Talk | Mapping Spiking Neural Networks onto a Manycore Neuromorphic Architecture PLDI Research Papers Chit-Kwan LinIntel Labs, n.n., Andreas WildIntel Labs, n.n., Tsung-Han LinIntel Labs, n.n., Gautham N. ChinyaIntel Labs, n.n., Mike DaviesIntel Labs, n.n., Hong WangIntel Labs, n.n. Media Attached |