Federated Few-Shot Learning on Neuromorphic Hardware: An Empirical Study Across Physical Edge Nodes
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866908884280541184 |
|---|---|
| author | Motta, Steven Nanni, Gioele |
| author_facet | Motta, Steven Nanni, Gioele |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_13037 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Federated Few-Shot Learning on Neuromorphic Hardware: An Empirical Study Across Physical Edge Nodes Motta, Steven Nanni, Gioele Neural and Evolutionary Computing Distributed, Parallel, and Cluster Computing Machine Learning I.2.6; I.2.11; C.2.4 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. |
| title | Federated Few-Shot Learning on Neuromorphic Hardware: An Empirical Study Across Physical Edge Nodes |
| topic | Neural and Evolutionary Computing Distributed, Parallel, and Cluster Computing Machine Learning I.2.6; I.2.11; C.2.4 |
| url | https://arxiv.org/abs/2603.13037 |