Using the method of intracellular injection of horseradish peroxidase, the arborizations of several neurones with different functions in the segmental ganglia of the leech Whitmania pigra are investigated. Comparison ...Using the method of intracellular injection of horseradish peroxidase, the arborizations of several neurones with different functions in the segmental ganglia of the leech Whitmania pigra are investigated. Comparison is made in each kind of neurones in order to verify their typical pattern. Results show that the morphology of the axonal arbors of identified neurones has an accurate similarity within and among the individuals of the leech, and the axonal arborizations of the bilaterally identified neurones in the same ganglion have a stable symmetry. The possible reasons of the formation and its functional significance of the symmetrical nervous system are discussed.展开更多
基金Project supported by the National Natural Science Foundation of China.
文摘Using the method of intracellular injection of horseradish peroxidase, the arborizations of several neurones with different functions in the segmental ganglia of the leech Whitmania pigra are investigated. Comparison is made in each kind of neurones in order to verify their typical pattern. Results show that the morphology of the axonal arbors of identified neurones has an accurate similarity within and among the individuals of the leech, and the axonal arborizations of the bilaterally identified neurones in the same ganglion have a stable symmetry. The possible reasons of the formation and its functional significance of the symmetrical nervous system are discussed.
文摘针对传统的神经网络模型因超参数众多,在实验中比对最优参数组合效率低下导致误差较大和反应速度慢的问题。本文提出一种基于北方苍鹰优化(Northern Goshawk Optimization,NGO)算法和双向门控循环单元神经网络(Bidirectional Gated Recurrent Unit, Bi-GRU)的船舶轨迹预测模型NGO-Bi-GRU(Northern Goshawk Optimization Bidirectional Gated Recurrent Unit)。利用NGO对Bi-GRU模型的学习率、隐藏节点和正则化系数进行寻优,然后将寻优得到的网络超参数代入Bi-GRU进行船舶轨迹预测。将该模型与长短时记忆神经网络(Long Short Term Memory, LSTM)和门控循环单元神经网络模型(Gated Recurrent Unit, GRU)以及使用该算法优化的长短期神经网络模型进行实验对比,将均方误差、均方根误差、平均绝对误差作为评价标准。结果表明,NGO-Bi-GRU模型在经度和纬度预测上误差较小、精确度较高且数值波动更加稳定。
