基于动态尖峰序列解码的状态估计和预测:噪声,自适应和多感知集成State estimation and prediction based on dynamic spike train decoding: noise, adaptation, and multisensory integration |
|
课程网址: | http://videolectures.net/aispds08_meir_sep/ |
主讲教师: | Ron Meir |
开课单位: | 以色列理工学院 |
开课时间: | 2008-08-05 |
课程语种: | 英语 |
中文简介: | 在不确定的动态环境中,有机体或一般的病原体所面临的一个关键需求是对环境状态的实时估计和预测,据此可以选择有效的行动。在这项工作中,我们展示了一个代理如何使用一个简单的实时神经网络,接收噪声多传感器输入信号,以有效地解决这些任务。 |
课程简介: | A key requirement facing organisms, or agents in general, acting in uncertain dynamic environments is the real-time estimation and prediction of environmental states, based upon which effective actions can be selected. In this work we show how an agent may use a simple real time neural network, receiving noisy multisensory input signals, to solve these tasks effectively. |
关 键 词: | 噪声; 适应; 多传感器融合 |
课程来源: | 视频讲座网 |
最后编审: | 2019-12-21:lxf |
阅读次数: | 50 |