论文部分内容阅读
神经计算机是仿效人脑神经系统的工作原理而研制的计算机。我们知道,计算机由于具有数值运算、逻辑判断、存储记忆等近似于人脑的功能而被誉为电脑。虽然电脑在高速运算、准确判断等逻辑思维方面有着超人之处,但在形象思维方面却远不及人脑。因为计算机的工作方式是“串行处理”,即每做一件事总要先分解成许多步,然后再一步一步地做。而从事图像识别等与形象思维相关的工作,则需要采用“并行处理”。譬如人脑,大脑皮层约有120亿个神经元(神经细胞),成千上万个神经元联成神经网络,既分工又合作地完成某项工作。这种由许多神经元同时工作完成同一任务的工作方式,就是并行处理。两三岁
Nerve computers are computers that follow the working principles of the human brain and nervous system. We know that computers have been hailed as computers for their functions as numerical operations, logical judgment, storage memory, and so on. Although computers are superhuman in the field of logical thinking such as high-speed computing and accurate judgment, they are far from the human brain in terms of image thinking. Because the way computers work is “serial processing”, that is, doing one thing always involves breaking it down into many steps first and then doing it step by step. In image recognition and other work related to image thinking, you need to use “parallel processing.” For example, in the human brain, there are about 12 billion neurons (nerve cells) in the cerebral cortex. Thousands of neurons are connected to a neural network to accomplish a certain task both in terms of division of labor and cooperation. This kind of work done by many neurons working on the same task at the same time is handled in parallel. Two or three years old