The mathematics behind spiking neural networks

报告人：Johannes Schmidt-Hiber（荷兰特文特大学）

时  间：2026年5月5日16:00

地  点：海韵园实验楼S102

内容摘要:

Artificial neural networks are inspired by the functioning of the brain but differ in several key aspects. In biological neural networks, information is encoded in the spiking times of neurons. In this survey talk, we first address the expressiveness of spiking neural networks and derive a universal representation theorem. 

Furthermore, it is implausible that biological learning is based on gradient descent. This has prompted researchers to propose various biologically inspired learning procedures. However, these methods lack a solid theoretical foundation. While statistical theory for artificial neural networks has been developed over the past years, the aim now is to extend this theory to biological neural networks, as the future of AI is likely to draw even more inspiration from biology. We will explore the challenges and present some recent theoretical results.

Joint work with Niklas Dexheimer, Sascha Gaudlitz, Shayan Hundrieser, Insung Kong, and Philipp Tuchel.

个人简介：

Johannes Schmidt-Hieber教授，国际数理统计学会会士（IMS Fellow），2025 年国际基础科学大会前沿科学奖得主。曾获荷兰统计学界最高荣誉Van Dantzig奖，并主持欧洲研究委员会（ERC）200万欧元资助项目。其研究成果在统计学及机器学习顶级期刊（如Annals of Statistics, JMLR, Bernoulli）多次发表，并担任Annals of Statistics，Bernoulli，Information and Inference, Electronic Journal of Statistics等多个国际权威期刊副编辑。 Schmidt-Hieber教授代表性论文单篇引用次数逾1400次，为深度学习的统计理论分析做出了突出贡献。

联系人：陈俊彤