Stack Filters Threshold Architectures and Information Encoding in Neural Systems

神经系统中的堆栈过滤器阈值架构和信息编码

• 批准号: 8721333
• 负责人: Edward Coyle
• 金额: $ 22.4万
• 依托单位: Purdue Research Foundation
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 1988
• 资助国家: 美国
• 起止时间: 1988-08-01 至 1991-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=8721333&HistoricalAwards=false
• 关键词: Stack; Filters; Threshold; Architectures; Information

The PI has developed a neural-like system called stack filters, which have achieved high performance in fixed image processing and show evidence of being more manageable, in some ways, than conventional neural networks. This project will attempt to develop learning algorithms for these systems, which will hopefully take advantage of their unique properties. These algorithms will be tested in speech recognition applications, in collaboration with a researcher who has developed a new programmable hearing implant for the deaf.

PI开发了一种称为堆栈过滤器的类神经系统， 在固定图像处理和显示方面取得了很高的性能 在某些方面比传统的更容易管理的证据， 神经网络 这个项目将尝试发展学习 这些系统的算法，这将有望利用 他们独特的属性。 这些算法将在语音中进行测试 识别应用程序，与研究人员合作， 为聋人开发了一种新的可编程听力植入物。