arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2406.03046cs.NE
收藏

当脉冲神经网络遇见时间注意力图像解码与自适应脉冲神经元

When Spiking neural networks meet temporal attention image decoding and adaptive spiking neuron

Xuerui Qiu, Zheng Luan, Zhaorui Wang, Rui-Jie Zhu

AI总结:

针对SNN未充分利用时间信息及忽略自适应阈值的问题,提出基于时间注意力的图像解码方法(TAID)和自适应LIF神经元模型(ALIF),在图像生成质量及MNIST和CIFAR-10分类准确率上均达到SOTA。

摘要:

脉冲神经网络(SNNs)能够以生物学合理的方式编码和处理时间信息。然而,现有大多数基于SNN的图像任务方法并未充分利用该特性。此外,它们常忽略自适应阈值在脉冲神经元中的作用,而该阈值能增强神经元的动态行为与学习能力。为解决这些问题,我们提出了一种基于时间注意力(TAID)的图像解码新方法及自适应漏电积分触发(ALIF)神经元模型。该方法利用SNN输出的时间信息生成高质量图像,在Inception分数、Fréchet Inception Distance和Fréchet Autoencoder Distance指标上超越了现有最先进(SOTA)水平。此外,ALIF模型在MNIST(99.78%)和CIFAR-10(93.89%)数据集上取得了显著的分类准确率,证明了学习自适应阈值对脉冲神经元的有效性。代码发布于https://github.com/bollossom/ICLR_TINY_SNN。

英文摘要:

Spiking Neural Networks (SNNs) are capable of encoding and processing temporal information in a biologically plausible way. However, most existing SNN-based methods for image tasks do not fully exploit this feature. Moreover, they often overlook the role of adaptive threshold in spiking neurons, which can enhance their dynamic behavior and learning ability. To address these issues, we propose a novel method for image decoding based on temporal attention (TAID) and an adaptive Leaky-Integrate-and-Fire (ALIF) neuron model. Our method leverages the temporal information of SNN outputs to generate high-quality images that surpass the state-of-the-art (SOTA) in terms of Inception score, Fréchet Inception Distance, and Fréchet Autoencoder Distance. Furthermore, our ALIF neuron model achieves remarkable classification accuracy on MNIST (99.78\%) and CIFAR-10 (93.89\%) datasets, demonstrating the effectiveness of learning adaptive thresholds for spiking neurons. The code is available at https://github.com/bollossom/ICLR_TINY_SNN.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑