Hyperdimensional Decoding of Spiking Neural Networks Publication type: Paper
Cedrick Kinavuidi 1,∗ \orcid 0009-0001-2860-9163, Luca Peres 1 \orcid 0000-0001-9748-9073 and Oliver Rhodes 1 \orcid 0000-0003-1728-2828 · 2025
Read Hyperdimensional Decoding of Spiking Neural Networks Publication type: Paper by Cedrick Kinavuidi 1,∗ \orcid 0009-0001-2860-9163, Luca Peres 1 \orcid 0000-0001-9748-9073 and Oliver Rhodes 1 \orcid 0000-0003-1728-2828 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-08558.txt.
Description
Cedrick Kinavuidi, Luca Peres, and Oliver Rhodes pair spiking neural networks with hyperdimensional computing to read out network activity. Spiking networks promise low energy use, but standard readouts based on firing rates or spike timing trade away accuracy, speed, or both.
On gesture and animal-action datasets recorded with event cameras, their SNN-HDC model classifies more accurately and faster than comparable designs. Estimated energy use drops by factors ranging from just over one to more than three, depending on the dataset and baseline. The decoder can also flag samples from a class it never saw in training. On one dataset it caught every unseen-class sample, a useful safety trait for open-world sensing.
Catalog record
| CREATOR | Cedrick Kinavuidi 1,∗ \orcid 0009-0001-2860-9163, Luca Peres 1 \orcid 0000-0001-9748-9073 and Oliver Rhodes 1 \orcid 0000-0003-1728-2828 |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 69 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2511-08558