MATHEMATICAL ANALYSES OF NEURAL NETWORKS

神经网络的数学分析

• 批准号: 3255955
• 负责人: BRUCE W KNIGHT
• 金额: $ 9.74万
• 依托单位: ROCKEFELLER UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 1977
• 资助国家: 美国
• 起止时间: 1977-09-01 至 1988-08-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/3255955
• 关键词: color visions; electrophysiology; mathematical model; neural transmission; neurons; psychophysics; retinal ganglion; saltwater environment; sequential perception; visual feedback; visual pathways; visual stimulus

We plan to continue our development of quantitative theoretical methods for study of the dynamics of the visual nervous system. These methods are a blend of modern applied analysis and computer techniques. We will continue to develop these procedures in close conjunction with experimental work. As applied to the eye of the horseshoe crab our theoretical methodology has reached a mature state; our intention is to extend it to vertebrate visual systems. Our approach involves three parts: The reduction of laboratory data, the use of those reduced data to formulate dynamical equations which summarize the behavior of the piece of nervous system under investigation, and the deduction of consequences from those equations to be compared against further experiments.

我们计划继续开发定量理论方法 研究视觉神经系统的动力学。 这些方法是 现代应用分析和计算机技术的混合物。 我们将继续 与实验工作密切结合开发这些程序。 正如适用于马蹄蟹的眼睛一样，我们的理论方法具有 达到一个成熟的状态；我们的目的是将其扩展到脊椎动物视觉 系统。 我们的方法涉及三个部分：减少实验室 数据，使用这些简化数据来制定动态方程 总结了正在调查的神经系统的行为， 并从这些方程式中扣除后果 反对进一步的实验。