Metraipolis › archive › dc_e70fc0605bd7

SpikeQ: Spike-Guided Post-Training Quantization for Spiking Neural Networks — arXiv:2610.23046

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.23046
License (as archived): CC0-1.0
Archived: 2026-10-19T08:43:07.311Z
SHA-256 of the archived text: 9393c1e4f4330027e767bce3481d54a9e6b5c8beebeb532cb3cc9f6d54dd9cd7
Stable URL of this page: https://metraipolis.com/documento/dc_e70fc0605bd7

How to cite

SpikeQ: Spike-Guided Post-Training Quantization for Spiking Neural Networks — arXiv:2610.23046 — arXiv (en), https://arxiv.org/abs/2610.23046 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e70fc0605bd7 (Metraipolis archive dc_e70fc0605bd7, sha256 9393c1e4f4330027e767bce3481d54a9e6b5c8beebeb532cb3cc9f6d54dd9cd7)

BibTeX

@misc{dce70fc0605bd7,
  title = {SpikeQ: Spike-Guided Post-Training Quantization for Spiking Neural Networks — arXiv:2610.23046},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.23046},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e70fc0605bd7; sha256 9393c1e4f4330027e767bce3481d54a9e6b5c8beebeb532cb3cc9f6d54dd9cd7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.23046. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

On neuromorphic chips, spiking neural networks (SNNs) consume far less energy than artificial neural networks (ANNs), but the limited memory of these chips requires low-bit synaptic weights and membrane potentials. Existing post-training quantization (PTQ) methods provide them without retraining and, like those for ANNs, minimize numerical errors such as the mean squared error (MSE). In an SNN, however, a quantization error reaches later layers only through the spikes it flips. We find that errors of equal MSE can cause very different accuracy losses, whereas errors that flip the same fraction of spikes cause the same loss. We therefore propose SpikeQ, a spike-guided PTQ framework. For weights, we select the level of each weight by simulating the downstream neurons and minimizing their spike errors. For membrane potentials, most existing methods quantize the state after the reset, whereas neuromorphic chips compare the threshold with the membrane potential stored at a fixed bit width. We target this pre-fire quantization and build a reset-aware grid that places the threshold on a rounding boundary, so rounding never flips the spike at the current time step. We evaluate SpikeQ on seven benchmarks of image classification, semantic segmentation, and object detection. With 2-bit pre-fire membrane potentials, every existing grid collapses, whereas SpikeQ retains most of the full-precision accuracy. With both at 4 bits, the models stay within 2.59 points of full precision, shrink their memory footprint 7.8--7.9 times, and use 6.4--6.9 times less energy.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Neural and Evolutionary Computing (cs.NE) ; Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2610.23046 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-19T08:39:10.139Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).