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 2406.16453q-bio.NCcond-mat.dis-nncond-mat.stat-mechcs.AIcs.NE
收藏

集合种群 Wilson-Cowan 模型中的学习

Learning in Wilson-Cowan model for metapopulation

University of Florence(佛罗伦萨大学)

Raffaele Marino, Lorenzo Buffoni, Lorenzo Chicchi, Francesca Di Patti, Diego Febbe, Lorenzo Giambagli, Duccio Fanelli

AI总结:

本文通过在集合种群 Wilson-Cowan 神经质量网络中引入稳定吸引子,将其改造为学习算法,并在多个图像与文本数据集上验证了高分类性能。

摘要:

用于集合种群的 Wilson-Cowan 模型是一种神经质量网络模型,它将大脑的不同皮层下区域视为相互连接的节点,而连接表示这些区域之间各种类型的结构性、功能性或有效性神经元连接。与标准 Wilson-Cowan 模型一致,每个区域都由相互作用的兴奋性细胞群和抑制性细胞群组成。通过将稳定吸引子纳入这种集合种群模型的动力学,我们将其转化为一种学习算法,能够实现较高的图像和文本分类准确率。我们在 MNIST 和 Fashion MNIST 上对其进行测试;结合卷积神经网络,在 CIFAR-10 和 TF-FLOWERS 上测试;并结合 transformer 架构(BERT),在 IMDB 上测试,结果始终表现出很高的分类准确率。这些数值评估表明,对集合种群 Wilson-Cowan 模型进行最小限度的修改,就可以揭示独特且此前未被观察到的动力学。

英文摘要:

The Wilson-Cowan model for metapopulation, a Neural Mass Network Model, treats different subcortical regions of the brain as connected nodes, with connections representing various types of structural, functional, or effective neuronal connectivity between these regions. Each region comprises interacting populations of excitatory and inhibitory cells, consistent with the standard Wilson-Cowan model. By incorporating stable attractors into such a metapopulation model's dynamics, we transform it into a learning algorithm capable of achieving high image and text classification accuracy. We test it on MNIST and Fashion MNIST, in combination with convolutional neural networks, on CIFAR-10 and TF-FLOWERS, and, in combination with a transformer architecture (BERT), on IMDB, always showing high classification accuracy. These numerical evaluations illustrate that minimal modifications to the Wilson-Cowan model for metapopulation can reveal unique and previously unobserved dynamics.

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。

↑