Application-oriented automatic hyperparameter optimization for spiking neural network prototyping

Vittorio Fra · 2025

Read Application-oriented automatic hyperparameter optimization for spiking neural network prototyping by Vittorio Fra (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-2502-12172.txt.

Description

Vittorio Fra, writing alone in 2025, presented an automatic pipeline for tuning spiking neural networks. Such networks use neuron-like units that fire in pulses, which adds extra settings beyond normal deep learning. Poor settings can hurt performance badly, so tuning matters a lot.

Fra uses the Neural Network Intelligence toolkit as the base for the pipeline. The approach is application-oriented: rather than chase a general model, it runs extensive tuning for the target use and accepts a possible loss of wider fit. A use case shows the pipeline in action and its results. A summary of past works that used the same pipeline gives further examples for builders of spiking network prototypes.

Catalog record

CREATORVittorio Fra
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE34 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2502-12172

Browse the full library · About the Tower of Babel