Accelerated physical emulation of Bayesian inference in spiking neural networks

Abstract

The massively parallel nature of biological information processing plays an important role for its superiority to human-engineered computing devices. In particular, it may hold the key to overcoming the von Neumann bottleneck that limits contemporary computer architectures. Physical-model neuromorphic devices seek to replicate not only this inherent parallelism, but also aspects of its microscopic dynamics in analog circuits emulating neurons and synapses. However, these machines require network models that are not only adept at solving particular tasks, but that can also cope with the inherent imperfections of analog substrates. We present a spiking network model that performs Bayesian inference through sampling on the BrainScaleS neuromorphic platform, where we use it for generative and discriminative computations on visual data. By illustrating its functionality on this platform, we implicitly demonstrate its robustness to various substrate-specific distortive effects, as well as its accelerated capability for computation. These results showcase the advantages of brain-inspired physical computation and provide important building blocks for large-scale neuromorphic applications.

arXiv commentThis preprint has been published 2019 November 14. Please cite as: Kungl A. F. et al. (2019) Accelerated Physical Emulation of Bayesian Inference in Spiking Neural Networks. Front. Neurosci. 13:1201. doi: 10.3389/fnins.2019.01201

References in corpus (21)

• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
• Solving the Quantum Many-Body Problem with Artificial Neural Networks
• Convolutional Networks for Fast, Energy-Efficient Neuromorphic Computing
• Decorrelation of neural-network activity by inhibitory feedback
• Better Mixing via Deep Representations
• Six networks on a universal neuromorphic computing substrate
• Demonstrating Advantages of Neuromorphic Computation: A Pilot Study
• Neuromorphic Hardware In The Loop: Training a Deep Spiking Network on the BrainScaleS Wafer-Scale System
• Quenches near Ising quantum criticality as a challenge for artificial neural networks
• Stochastic inference with spiking neurons in the high-conductance state
• Characterization and Compensation of Network-Level Anomalies in Mixed-Signal Neuromorphic Modeling Platforms
• Stochasticity from function -- why the Bayesian brain may need no noise
• Stochastic inference with deterministic spiking neurons
• Deterministic networks for probabilistic computing
• Pattern representation and recognition with accelerated analog neuromorphic systems
• Mapping Generative Models onto a Network of Digital Spiking Neurons
• Probabilistic inference in discrete spaces can be implemented into networks of LIF neurons
• The effect of heterogeneity on decorrelation mechanisms in spiking neural networks: a neuromorphic-hardware study
• Spike-based probabilistic inference with correlated noise
• Fast sampling for Bayesian inference in neural circuits
• The Neural Particle Filter

Cited by in corpus (9)

• Differentiable graph-structured models for inverse design of lattice materials
• Cortical oscillations implement a backbone for sampling-based computation in spiking neural networks
• Three Learning Stages and Accuracy-Efficiency Tradeoff of Restricted Boltzmann Machines
• Spiking neuromorphic chip learns entangled quantum states
• Automotive Object Detection via Learning Sparse Events by Spiking Neurons
• Sampling scheme for neuromorphic simulation of entangled quantum systems
• Fast and energy-efficient neuromorphic deep learning with first-spike times
• Efficient Deep Spiking Multi-Layer Perceptrons with Multiplication-Free Inference
• Ultrafast neural sampling with spiking nanolasers