Comparing Loihi with a SpiNNaker 2 prototype on low-latency keyword spotting and adaptive robotic control

We implemented two neural network based benchmark tasks on a prototype chip of the second-generation SpiNNaker (SpiNNaker 2) neuromorphic system: keyword spotting and adaptive robotic control. Keyword spotting is commonly used in smart speakers to listen for wake words, and adaptive control is used in robotic applications to adapt to unknown dynamics in an online fashion. We highlight the benefit of a multiply-accumulate (MAC) array in the SpiNNaker 2 prototype which is ordinarily used in rate-based machine learning networks when employed in a neuromorphic, spiking context. In addition, the same benchmark tasks have been implemented on the Loihi neuromorphic chip, giving a side-by-side comparison regarding power consumption and computation time. While Loihi shows better efficiency when less complicated vector-matrix multiplication is involved, with the MAC array, the SpiNNaker 2 prototype shows better efficiency when high dimensional vector-matrix multiplication is involved.

[1]  Steve B. Furber,et al.  Efficient Reward-Based Structural Plasticity on a SpiNNaker 2 Prototype , 2019, IEEE Transactions on Biomedical Circuits and Systems.

[2]  Kaushik Roy,et al.  Towards spike-based machine intelligence with neuromorphic computing , 2019, Nature.

[3]  Lakmal Seneviratne,et al.  Adaptive Control Of Robot Manipulators , 1992, Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems.

[4]  Garrick Orchard,et al.  Online Few-Shot Gesture Learning on a Neuromorphic Processor , 2020, IEEE Journal on Emerging and Selected Topics in Circuits and Systems.

[5]  Gert Cauwenberghs,et al.  Hierarchical Address Event Routing for Reconfigurable Large-Scale Neuromorphic Systems , 2017, IEEE Transactions on Neural Networks and Learning Systems.

[6]  Christian Mayr,et al.  True random number generation from bang-bang ADPLL jitter , 2016, 2016 IEEE Nordic Circuits and Systems Conference (NORCAS).

[7]  Giacomo Indiveri,et al.  A reconfigurable on-line learning spiking neuromorphic processor comprising 256 neurons and 128K synapses , 2015, Front. Neurosci..

[8]  Nabil Imam,et al.  Rapid online learning and robust recall in a neuromorphic olfactory circuit , 2019, Nature Machine Intelligence.

[9]  Johannes Schemmel,et al.  Demonstrating Hybrid Learning in a Flexible Neuromorphic Hardware System , 2016, IEEE Transactions on Biomedical Circuits and Systems.

[10]  Steve B. Furber,et al.  The SpiNNaker Project , 2014, Proceedings of the IEEE.

[11]  Terrence C. Stewart,et al.  Closed-Loop Neuromorphic Benchmarks , 2015, Front. Neurosci..

[12]  Rodrigo Alvarez-Icaza,et al.  Neurogrid: A Mixed-Analog-Digital Multichip System for Large-Scale Neural Simulations , 2014, Proceedings of the IEEE.

[13]  René Schüffny,et al.  A Biological-Realtime Neuromorphic System in 28 nm CMOS Using Low-Leakage Switched Capacitor Circuits , 2014, IEEE Transactions on Biomedical Circuits and Systems.

[14]  Trevor Bekolay,et al.  A Large-Scale Model of the Functioning Brain , 2012, Science.

[15]  Vivienne Sze,et al.  Efficient Processing of Deep Neural Networks: A Tutorial and Survey , 2017, Proceedings of the IEEE.

[16]  Steve B. Furber,et al.  A fixed point exponential function accelerator for a neuromorphic many-core system , 2017, 2017 IEEE International Symposium on Circuits and Systems (ISCAS).

[17]  Mingguo Zhao,et al.  Towards artificial general intelligence with hybrid Tianjic chip architecture , 2019, Nature.

[18]  Steve Furber,et al.  Dynamic Power Management for Neuromorphic Many-Core Systems , 2019, IEEE Transactions on Circuits and Systems I: Regular Papers.

[19]  Alois Knoll,et al.  The Human Brain Project: Creating a European Research Infrastructure to Decode the Human Brain , 2016, Neuron.

[20]  Kwabena Boahen,et al.  Braindrop: A Mixed-Signal Neuromorphic Architecture With a Dynamical Systems-Based Programming Model , 2019, Proceedings of the IEEE.

[21]  Steve Furber,et al.  SpiNNaker 2: A 10 Million Core Processor System for Brain Simulation and Machine Learning , 2019, ArXiv.

[22]  H. Markram,et al.  Regulation of Synaptic Efficacy by Coincidence of Postsynaptic APs and EPSPs , 1997, Science.

[23]  Andrew S. Cassidy,et al.  A million spiking-neuron integrated circuit with a scalable communication network and interface , 2014, Science.

[24]  Jim D. Garside,et al.  Approximate Fixed-Point Elementary Function Accelerator for the SpiNNaker-2 Neuromorphic Chip , 2018, 2018 IEEE 25th Symposium on Computer Arithmetic (ARITH).

[25]  Chris Eliasmith,et al.  Neural Engineering: Computation, Representation, and Dynamics in Neurobiological Systems , 2004, IEEE Transactions on Neural Networks.

[26]  Chris Eliasmith,et al.  A spiking neural model of adaptive arm control , 2016, Proceedings of the Royal Society B: Biological Sciences.

[27]  Stephan Henker,et al.  Highly integrated packet-based AER communication infrastructure with 3Gevent/S throughput , 2010, 2010 17th IEEE International Conference on Electronics, Circuits and Systems.

[28]  Stephen B. Furber,et al.  An efficient SpiNNaker implementation of the Neural Engineering Framework , 2015, 2015 International Joint Conference on Neural Networks (IJCNN).

[29]  Walter Senn,et al.  Learning Real-World Stimuli in a Neural Network with Spike-Driven Synaptic Dynamics , 2007, Neural Computation.

[30]  Dennis Walter,et al.  A 0.55V 6.3uW/MHz Arm Cortex-M4 MCU with Adaptive Reverse Body Bias and Single Rail SRAM , 2020, 2020 IEEE Symposium in Low-Power and High-Speed Chips (COOL CHIPS).

[31]  Dennis Walter,et al.  How to Achieve World-Leading Energy Efficiency using 22FDX with Adaptive Body Biasing on an Arm Cortex-M4 IoT SoC , 2019, ESSDERC 2019 - 49th European Solid-State Device Research Conference (ESSDERC).

[32]  C. Eliasmith,et al.  Nengo and Low-Power AI Hardware for Robust, Embedded Neurorobotics , 2020, Frontiers in Neurorobotics.

[33]  Steve B. Furber,et al.  Power analysis of large-scale, real-time neural networks on SpiNNaker , 2013, The 2013 International Joint Conference on Neural Networks (IJCNN).

[34]  Chris Eliasmith,et al.  Fine-Tuning and the Stability of Recurrent Neural Networks , 2011, PloS one.

[35]  Jim D. Garside,et al.  Dynamic voltage and frequency scaling for neuromorphic many-core systems , 2017, 2017 IEEE International Symposium on Circuits and Systems (ISCAS).

[36]  David Bol,et al.  MorphIC: A 65-nm 738k-Synapse/mm$^2$ Quad-Core Binary-Weight Digital Neuromorphic Processor With Stochastic Spike-Driven Online Learning , 2019, IEEE Transactions on Biomedical Circuits and Systems.

[37]  J. Slotine,et al.  On the Adaptive Control of Robot Manipulators , 1987 .

[38]  Steve B. Furber,et al.  Efficient SpiNNaker simulation of a heteroassociative memory using the Neural Engineering Framework , 2016, 2016 International Joint Conference on Neural Networks (IJCNN).

[39]  D. Johnston,et al.  Regulation of Synaptic Efficacy by Coincidence of Postsynaptic APs and EPSPs , 1997 .

[40]  Jim D. Garside,et al.  The SpiNNaker 2 Processing Element Architecture for Hybrid Digital Neuromorphic Computing , 2021, ArXiv.

[41]  Hong Wang,et al.  Loihi: A Neuromorphic Manycore Processor with On-Chip Learning , 2018, IEEE Micro.

[42]  André van Schaik,et al.  An FPGA-Based Massively Parallel Neuromorphic Cortex Simulator , 2018, Front. Neurosci..

[43]  Stephan Henker,et al.  Adaptive Body Bias Aware Implementation for Ultra-Low-Voltage Designs in 22FDX Technology , 2020, IEEE Transactions on Circuits and Systems II: Express Briefs.

[44]  Giacomo Indiveri,et al.  A Scalable Multicore Architecture With Heterogeneous Memory Structures for Dynamic Neuromorphic Asynchronous Processors (DYNAPs) , 2017, IEEE Transactions on Biomedical Circuits and Systems.

[45]  Peter Blouw,et al.  Benchmarking Keyword Spotting Efficiency on Neuromorphic Hardware , 2018, NICE '19.

[46]  Massimiliano Giulioni,et al.  Mean Field Approach for Configuring Population Dynamics on a Biohybrid Neuromorphic System , 2019, Journal of Signal Processing Systems.

[47]  David Bol,et al.  A 0.086-mm2 12.7-pJ/SOP 64k-Synapse 256-Neuron Online-Learning Digital Spiking Neuromorphic Processor in 28-nm CMOS , 2019, IEEE Trans. Biomed. Circuits Syst..