Skip to content
Hot eventLive

Controllable Stochastic Quantization Encoding for Adversarially Robust Spiking Neural Networks

1 reports1 sources3 hr ago updated

Get the story

AI overview

2026-10-02,arXiv 发表一篇 Neural and Evolutionary Computing 类别的论文,提出可控随机量化编码方法,通过调节量化尺度为输入图像注入可控随机性,以提升脉冲神经网络(SNN)的对抗鲁棒性。该方法在不同量化尺度下可退化为泊松编码与直接编码,构成通用框架,并可与现有基于训练的防御方法结合增强鲁棒性。实验在 CIFAR-10 与 CIFAR-100 上验证了有效性。

Generated from reports · updated 2 hr ago

Timeline

Follow the coverage from different angles.

Oct 2, 2026
  1. arXiv · Neural and Evolutionary Computing
    Controllable Stochastic Quantization Encoding for Adversarially Robust Spiking Neural Networks

    论文提出可控随机量化编码方法,通过调节量化尺度为输入图像注入可控随机性,提升脉冲神经网络(SNN)的对抗鲁棒性。该方法在不同量化尺度下可退化为泊松编码与直接编码,构成通用框架,并可与现有基于训练的防御方法结合进一步增强鲁棒性。实验在 CIFAR-10 与 CIFAR-100 上验证了有效性。

Heat trend

There is not enough continuous observation data to draw a trend yet.