Identification of motion with echo state network
Identification of motion with echo state network
复制标题
DOI：
10.1109/oceans.2004.1405751
复制
发表时间：
2004-11
期刊：
影响因子：
--
通讯作者：
K. Ishu;T. V. D. Zant;V. Becanovic;P. Plöger
中科院分区：
文献类型：
--
作者：
K. Ishu;T. V. D. Zant;V. Becanovic;P. Plöger
回声状态网络（ESN）使用递归人工神经网络作为水库。找到一个好的取决于选择正确的参数来生成储层，直觉和运气。本文中提出的方法通过用双重进化计算代替手工调整来消除调整的需要。首先，使用广泛搜索以找到生成储层的正确参数。然后直接在连接矩阵上进行搜索来微调ESN。这两个步骤都显示了对Twin-Burger水下机器人实验极限环数据集的其他已知方法的改进。
Echo State Networks (ESNs) use a recurrent artificial neural network as a reservoir. Finding a good one depends on choosing the right parameters for the generation of the reservoir, intuition and luck. The method proposed in this article eliminates the need for the tuning by hand by replacing it with a double evolutionary computation. First a broad search to find the right parameters which generate the reservoir is used. Then a search directly on the connectivity matrices fine-tunes the ESN. Both steps show improvements over other known methods for an experimental limit-cycle dataset of the Twin-Burger underwater robot.