Neuromorphic Hardware In The Loop: Training a Deep Spiking Network on the BrainScaleS Wafer-Scale System

Abstract

Emulating spiking neural networks on analog neuromorphic hardware offers several advantages over simulating them on conventional computers, particularly in terms of speed and energy consumption. However, this usually comes at the cost of reduced control over the dynamics of the emulated networks. In this paper, we demonstrate how iterative training of a hardware-emulated network can compensate for anomalies induced by the analog substrate. We first convert a deep neural network trained in software to a spiking network on the BrainScaleS wafer-scale neuromorphic system, thereby enabling an acceleration factor of 10 000 compared to the biological time domain. This mapping is followed by the in-the-loop training, where in each training step, the network activity is first recorded in hardware and then used to compute the parameter updates in software via backpropagation. An essential finding is that the parameter updates do not have to be precise, but only need to approximately follow the correct gradient, which simplifies the computation of updates. Using this approach, after only several tens of iterations, the spiking network shows an accuracy close to the ideal software-emulated prototype. The presented techniques show that deep spiking networks emulated on analog neuromorphic devices can attain good computational performance despite the inherent variations of the analog substrate.

8 pages, 10 figures, submitted to IJCNN 2017

Cited by in corpus (24)

• Brain-inspired computing: We need a master plan
• Hardware and Software Optimizations for Accelerating Deep Neural Networks: Survey of Current Trends, Challenges, and the Road Ahead
• Neuromorphic Intermediate Representation: A Unified Instruction Set for Interoperable Brain-Inspired Computing
• A Mixed-Signal Structured AdEx Neuron for Accelerated Neuromorphic Cores
• Versatile emulation of spiking neural networks on an accelerated neuromorphic substrate
• Accelerated physical emulation of Bayesian inference in spiking neural networks
• RANC: Reconfigurable Architecture for Neuromorphic Computing
• Efficient Reward-Based Structural Plasticity on a SpiNNaker 2 Prototype
• Pattern representation and recognition with accelerated analog neuromorphic systems
• Integration of Neuromorphic AI in Event-Driven Distributed Digitized Systems: Concepts and Research Directions
• Spiking Neural Network Nonlinear Demapping on Neuromorphic Hardware for IM/DD Optical Communication
• Demonstrating Analog Inference on the BrainScaleS-2 Mobile System
• hxtorch: PyTorch for BrainScaleS-2 -- Perceptrons on Analog Neuromorphic Hardware
• NeuroPack: An Algorithm-level Python-based Simulator for Memristor-empowered Neuro-inspired Computing
• Scaling of a large-scale simulation of synchronous slow-wave and asynchronous awake-like activity of a cortical model with long-range interconnections
• Fast and energy-efficient neuromorphic deep learning with first-spike times
• An Inference and Learning Engine for Spiking Neural Networks in Computational RAM (CRAM)
• Inference with Artificial Neural Networks on Analog Neuromorphic Hardware
• Gaussian and exponential lateral connectivity on distributed spiking neural network simulation
• Distributed Representations Enable Robust Multi-Timescale Symbolic Computation in Neuromorphic Hardware
• Benchmarking Deep Spiking Neural Networks on Neuromorphic Hardware
• D-SELD: Dataset-Scalable Exemplar LCA-Decoder
• Demonstrating BrainScaleS-2 Inter-Chip Pulse-Communication using EXTOLL
• Exploring vestibulo-ocular adaptation in a closed-loop neuro-robotic experiment using STDP. A simulation study