Scholarlink

The decision-making of artificial neural networks, especially in Spiking Neural Networks (SNNs) has always attracted researchers’ attention. Based on the concept of evidence accumulation, the Drift Diffusion Model is adopted in this work to make the proposed SNN have the ability of autonomous decision-making. The proposed SNN is trained by the constructed reward modulation learning algorithm based on the combination of the reinforcement learning and the Drift Diffusion Model. In the end, the proposed network model and the learning algorithm are verified by the XOR experiment. The experimental results demonstrate that the proposed neural network model has the ability of autonomous decision-making.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Autonomous Decision-Making Spiking Neural Network Based on Evidence Accumulation

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Junxiu Liu,
  • Hao Lu,
  • Yuling Luo,
  • Qiang Fu,
  • Xue Ouyang,
  • Sheng Qin,
  • Changyong Yang

摘要

The decision-making of artificial neural networks, especially in Spiking Neural Networks (SNNs) has always attracted researchers’ attention. Based on the concept of evidence accumulation, the Drift Diffusion Model is adopted in this work to make the proposed SNN have the ability of autonomous decision-making. The proposed SNN is trained by the constructed reward modulation learning algorithm based on the combination of the reinforcement learning and the Drift Diffusion Model. In the end, the proposed network model and the learning algorithm are verified by the XOR experiment. The experimental results demonstrate that the proposed neural network model has the ability of autonomous decision-making.