Research

Paper

TESTING March 13, 2026

Federated Few-Shot Learning on Neuromorphic Hardware: An Empirical Study Across Physical Edge Nodes

Authors

Steven Motta, Gioele Nanni

Abstract

Federated learning on neuromorphic hardware remains unexplored because on-chip spike-timing-dependent plasticity (STDP) produces binary weight updates rather than the floating-point gradients assumed by standard algorithms. We build a two-node federated system with BrainChip Akida AKD1000 processors and run approximately 1,580 experimental trials across seven analysis phases. Of four weight-exchange strategies tested, neuron-level concatenation (FedUnion) consistently preserves accuracy while element-wise weight averaging (FedAvg) destroys it (p = 0.002). Domain-adaptive fine-tuning of the upstream feature extractor accounts for most of the accuracy gains, confirming feature quality as the dominant factor. Scaling feature dimensionality from 64 to 256 yields 77.0% best-strategy federated accuracy (n=30, p < 0.001). Two independent asymmetries (wider features help federation more than individual learning, while binarization hurts federation more) point to a shared prototype complementarity mechanism: cross-node transfer scales with the distinctiveness of neuron prototypes.

Metadata

arXiv ID: 2603.13037
Provider: ARXIV
Primary Category: cs.NE
Published: 2026-03-13
Fetched: 2026-03-16 06:01

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.13037v1</id>\n    <title>Federated Few-Shot Learning on Neuromorphic Hardware: An Empirical Study Across Physical Edge Nodes</title>\n    <updated>2026-03-13T14:45:52Z</updated>\n    <link href='https://arxiv.org/abs/2603.13037v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.13037v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Federated learning on neuromorphic hardware remains unexplored because on-chip spike-timing-dependent plasticity (STDP) produces binary weight updates rather than the floating-point gradients assumed by standard algorithms. We build a two-node federated system with BrainChip Akida AKD1000 processors and run approximately 1,580 experimental trials across seven analysis phases. Of four weight-exchange strategies tested, neuron-level concatenation (FedUnion) consistently preserves accuracy while element-wise weight averaging (FedAvg) destroys it (p = 0.002). Domain-adaptive fine-tuning of the upstream feature extractor accounts for most of the accuracy gains, confirming feature quality as the dominant factor. Scaling feature dimensionality from 64 to 256 yields 77.0% best-strategy federated accuracy (n=30, p &lt; 0.001). Two independent asymmetries (wider features help federation more than individual learning, while binarization hurts federation more) point to a shared prototype complementarity mechanism: cross-node transfer scales with the distinctiveness of neuron prototypes.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.NE'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.DC'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.LG'/>\n    <published>2026-03-13T14:45:52Z</published>\n    <arxiv:comment>13 pages, 2 figures, 10 tables. Code: https://github.com/Stemo688/federated-neuromorphic-learning</arxiv:comment>\n    <arxiv:primary_category term='cs.NE'/>\n    <author>\n      <name>Steven Motta</name>\n    </author>\n    <author>\n      <name>Gioele Nanni</name>\n    </author>\n  </entry>"
}