We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: Federated Learning with Spiking Neural Networks

Abstract: As neural networks get widespread adoption in resource-constrained embedded devices, there is a growing need for low-power neural systems. Spiking Neural Networks (SNNs)are emerging to be an energy-efficient alternative to the traditional Artificial Neural Networks (ANNs) which are known to be computationally intensive. From an application perspective, as federated learning involves multiple energy-constrained devices, there is a huge scope to leverage energy efficiency provided by SNNs. Despite its importance, there has been little attention on training SNNs on a large-scale distributed system like federated learning. In this paper, we bring SNNs to a more realistic federated learning scenario. Specifically, we propose a federated learning framework for decentralized and privacy-preserving training of SNNs. To validate the proposed federated learning framework, we experimentally evaluate the advantages of SNNs on various aspects of federated learning with CIFAR10 and CIFAR100 benchmarks. We observe that SNNs outperform ANNs in terms of overall accuracy by over 15% when the data is distributed across a large number of clients in the federation while providing up to5.3x energy efficiency. In addition to efficiency, we also analyze the sensitivity of the proposed federated SNN framework to data distribution among the clients, stragglers, and gradient noise and perform a comprehensive comparison with ANNs.
Subjects: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV); Distributed, Parallel, and Cluster Computing (cs.DC); Neural and Evolutionary Computing (cs.NE)
DOI: 10.1109/TSP.2021.3121632
Cite as: arXiv:2106.06579 [cs.LG]
  (or arXiv:2106.06579v1 [cs.LG] for this version)

Submission history

From: Yeshwanth Venkatesha [view email]
[v1] Fri, 11 Jun 2021 19:00:58 GMT (30905kb,D)

Link back to: arXiv, form interface, contact.