The Heidelberg Spiking Data Sets for the Systematic Evaluation of Spiking Neural Networks
Хайдельбергские наборы данных о спайках для систематической оценки спайковых нейронных сетей
2020-12-30
SCID: 54.1/4vrrcj4r
Discuss with AI
Heidelberg Spiking Data Setsaudio-to-spiking conversionneuromorphic hardwarespike-based classificationspiking neural networks
Figures from the paper
Abstract (AI)
Spiking neural networks are the basis of versatile and power-efficient information processing in the brain. Although we currently lack a detailed understanding of how these networks compute, recently developed optimization techniques allow us to instantiate increasingly complex functional spiking neural networks in-silico. These methods hold the promise to build more efficient non-von-Neumann computing hardware and will offer new vistas in the quest of unraveling brain circuit function. To accelerate the development of such methods, objective ways to compare their performance are indispensable. Presently, however, there are no widely accepted means for comparing the computational performance of spiking neural networks. To address this issue, we introduce two spike-based classification data sets, broadly applicable to benchmark both software and neuromorphic hardware implementations of spiking neural networks. To accomplish this, we developed a general audio-to-spiking conversion procedure inspired by neurophysiology. Furthermore, we applied this conversion to an existing and a novel speech data set. The latter is the free, high-fidelity, and word-level aligned Heidelberg digit data set that we created specifically for this study. By training a range of conventional and spiking classifiers, we show that leveraging spike timing information within these data sets is essential for good classification accuracy. These results serve as the first reference for future performance comparisons of spiking neural networks.
Key Findings
1
Experiments with conventional and spiking classifiers show that exploiting spike timing information is essential for achieving good classification accuracy.
2
It presents a general audio-to-spiking conversion procedure inspired by neurophysiology for transforming speech signals into spike-based representations.
3
The authors create the free, high-fidelity, word-level aligned Heidelberg digit dataset specifically for evaluating spiking neural networks.
4
The reported results provide an initial reference point for comparing the computational performance of future spiking neural network methods.
5
The study introduces two spike-based classification datasets designed for systematic benchmarking of spiking neural networks in software and neuromorphic hardware.
Research Object
the Heidelberg spike-based classification datasets, including the Heidelberg digit dataset
Research Subject
the classification performance of spiking neural networks and the contribution of spike-timing information
Publication Details
Publication Date
2020-12-30
Journal
Publisher
ISSN
Cited by
351
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai6
Adam: A Method for Stochastic Optimization2014
Scikit-learn: Machine Learning in Python2012
Gradient-based learning applied to document recognition1998
Learning Multiple Layers of Features from Tiny Images2024
Surrogate Gradient Learning in Spiking Neural Networks: Bringing the Power of Gradient-Based Optimization to Spiking Neural Networks2019
SuperSpike: Supervised Learning in Multilayer Spiking Neural Networks2018