Training in Methods of Computational Neuroscience

描述（由申请人提供）：在本提案中，我们要求继续支持自1988年以来每年在海洋生物实验室提供的计算神经科学课程方法。本课程的目标是每年培训24名高级学生和博士后研究员，研究计算神经科学的基本问题，为他们提供分析工具，以协助他们的研究。该课程有几个独特的特点，这将是很难，如果不是不可能在大学环境中复制。首先，它继续吸引和教育来自广泛的影响神经科学的既定学科的学生，包括生物学，认知科学，计算机科学，数学，物理学和工程学。其次，作为一种手段，以提供一个连贯的和相对完整的角度对计算神经科学的不断增长的领域，学生接触到一个大型的访问教师，从该领域的领导者，以及接收指令和指导课程主任，驻地教师和课程助理绘制。这是不可能的，这样的教师组合可以召集类似的课程在大学设置。最后，该课程为学生提供了从详细的细胞到更抽象的系统水平的神经系统模拟中必不可少的最先进的计算技术和计算机硬件。该课程是一个密集的4周讲座/实验室系列。前三周的每一周都围绕一个主题进行组织，包括概念上的和开发的数学工具。第1周讲述神经编码。在第2周，我们将研究计算是如何从单个神经元和网络的动力学特性中产生的。第三周的重点是学习，并使用学习理论，估计和控制理论的方法。最后一周的大部分时间都用于个别项目。我们的目标是向学生展示实验和理论相结合的方法如何带来突破，并使他们熟悉成功所需的分析方法和计算工具。 公共卫生相关性：神经系统疾病可能是毁灭性的。了解这些疾病的基本原理最终可以导致治愈。神经科学中的计算方法开始揭示基于表现出周期性或周期性的神经精神和神经疾病，包括双相情感障碍、精神分裂症、季节性情感障碍、Klein-Levin综合征、睡眠障碍、暴食、癫痫、多发性硬化症、时差综合征和头痛。本提案中提出的计算神经科学方法课程是一个密集课程，由该领域的领导者讲授，致力于培训该领域的研究人员。该计划现已进入第22个年头，已经培养了许多熟练的研究人员，他们在神经系统疾病的研究中做出了重大贡献。目前提议的目的是为今后继续培训这类调查员，

DESCRIPTION (provided by applicant): In this proposal we request continuation of support for the Methods in Computational Neuroscience Course offered annually at the Marine Biological Laboratory since 1988. The goal of this course is to train 24 advanced students and post-doctoral fellows per year in the fundamental issues in computational neuroscience, providing them with analytical tools to assist in their studies. The course has several unique features, which would be hard if not impossible to duplicate in a university setting. First, it continues to attract and educate students from a broad range of established disciplines that impact on neuroscience, including biology, cognitive science, computer science, mathematics, physics and engineering. Secondly, as a means to providing a coherent and relatively complete perspective on the growing field of computational neuroscience, students are exposed to a large visiting faculty, drawn from leaders in the field, as well as receiving instruction and guidance from the course directors, resident faculty and course assistants. It is unlikely that such a combination of faculty could be convened for a similar course in a university setting. Finally, the course provides students with access to state-of-art computational techniques and computer hardware essential in the simulation of neural systems from a detailed cellular through more abstract systems level. The course is an intensive 4-week lecture/laboratory series. Each of the first three weeks is organized around a theme, both conceptually and in terms of the mathematical tools developed. Week 1 addresses neural coding. In Week 2, we examine how computation emerges from the dynamical properties of single neurons and networks. Week 3 focuses on learning and uses methods from learning theory, estimation and control theory. Most of the final week is devoted to individual projects. We aim to show students how combined experimental and theoretical approaches can lead to breakthroughs and make them familiar with the analytic methods and computational tools that they will need for success. PUBLIC HEALTH RELEVANCE: Diseases of the nervous system can be devastating. Understanding the fundamental basis for these diseases can eventually lead to cures. Computational approaches in neuroscience are beginning to shed light on the basis of neuropsychiatric and neurological diseases exhibiting periodicity or cyclicity which include Bipolar Disorder, Schizophrenia, Seasonal Affective Disorder, Klein-Levin Syndrome, Sleep Disorders, Binging, Epilepsy, Multiple Sclerosis, Jet-Lag and Headache. The course on Methods in Computational Neuroscience presented in this proposal is an intensive program, taught by leaders in the field, dedicated to training investigators in this area. This program, now entering its 22nd year, has already produced many skilled investigators who are making significant contributions in their studies on diseases of the nervous system. The aim of the current proposal is to continue training such investigators for the future,