通用神经网络硬件中神经元基本数学模型的讨论

Discussion on the Basic Mathematical Models of Neurons in General Purpose Neurocomputer

在介绍了作者实现通用神经网络硬件中应用的通用计算公式的基础上 ,提出了一种能同时模拟包括RBF与传统BP网络神经元在内的各种神经元通用的新的数学计算模型 ,并把基于这种通用数学计算模型的神经网络CASSANDRA Ⅱ型神经计算机结构设计中并予以硬件实现 .文中还讨论了它所模拟神经元网络的灵活性 .

Based on the introduction of the traditional mathematical model of neurons in general-purpose neural network hardware realization, a novel all-purpose mathematical model is proposed, which can simulate all kinds of neuron architecture, including RBF and BP models, etc. At the same time this model is realized using hardware and implemented in the new CASSANDRA-II neurocomputer that can be used to form many neural networks with multiple mathematical models of neurons. In this paper the flexibility of the new neurons is also discussed in constructing neural networks.