兴奋抑制混合神经网络的频率控制

The collaboration between the complex structure and the various functions on the brain is a hot topic. The dynamical complexity of the neurons play an important role. In the past two decades, a lot of progress has been developed on the experimental and theoretical works. Now the research interest is mainly focused on the super-large-scale structure in the neuronal system, the memory mechanism, the brain waves and so on. Several method for the low frequency exciting has been discussed in the previous theoretical and experimental works, and we will discuss one of them-the frequency match method. In this project, considering the various types of neurons, we will construct the network model mixed with the excitatory and inhibitory neurons from the experimental data, and observe the functional effect of the ratio of inhibitory neurons, frequency mismatch and dynamic effect of the local network circuits. The relation between phase coherence and frequency locking has been widely discussed in many oscillation models. We will also discuss the influence of the frequency mismatch and excitatory-inhibitory balance for the signal transmission and information coding. The result will improve the method of frequency control and help to understand the mechanism of information processing on the neuronal network.

大脑的复杂结构与其多样功能的对应是当今热门的课题。神经元自身动力学的丰富性与复杂性是其中的关键因素之一。过去20多年，对神经元局域回路、突触修饰、病态振子等的实验与理论工作都取得了相当的进展。如今研究人员的目光将更多集中与神经系统中的记忆机制研究，脑波研究,大尺度结构研究等方向。前人的理论研究与实验数据中，神经系统的低频激发机制有很多不同的达成方式，本项目则讨论其中之一的频率匹配方法。本项目将部分依托于前人实验所发现的海马回区域振子分类，而构造多样化的兴奋抑制混合神经网络，具体考察抑制振子比例、多种频率振子在网络中对频率控制与低频生成的共同作用，关注其网络局域结构对其动力学的影响。通过借鉴前人对振子系统相关联与锁频等研究，我们还将分析因为兴奋抑制的平衡、多种频率振子的混杂对信号传输以及信息编码的影响。这些讨论将有助于完善对神经网络频率控制的研究并帮助理解网络中信息处理的机制。

生命系统的复杂结构与其多样功能的对应是当今热门的课题。特别是神经系统自身动力学的丰富性依然需要长足的探索。本项目着重考察在生命网络特别是神经网络上结构与动力学功能间的关联，取得了部分成果。1.我们通过构造多样化的兴奋抑制混合神经网络，实现了频率控制，在此基础上我们考察了抑制振子比例、网络局域结构、振子耦合参数对其动力学的影响。2.通过小波修饰，我们完成了在神经网络结构上的特异性同步斑图，并据此讨论了神经元成长衰退机制。3. 通过对酵母的基因调控网的分析，我们给出了一种寻找最小功能骨架网的算法，并发现这一算法更加高效、普适。我们相信这些讨论将有助于对生命系统中复杂动力学的理解。