首页>
外国专利>
Neural Networks ENTROPY-BASED NEURAL NETWORKS PARTIAL LEARNING METHOD AND SYSTEM
Neural Networks ENTROPY-BASED NEURAL NETWORKS PARTIAL LEARNING METHOD AND SYSTEM
展开▼
机译:神经网络的基于熵的神经网络局部学习方法和系统
展开▼
页面导航
摘要
著录项
相似文献
摘要
The present invention relates to a method and a system for maintaining accuracy while reducing a load of learning when a new class appears in learning by using convolutional neural networks, and more particularly, to a partial learning method of convolutional neural networks by weight evaluation based on entropy and a system therefor. The learning method by using neural networks comprises the steps of: (a) recognizing the occurrence of a new class; (b) calculating a threshold value for determining qualitative information based on the entropy of a plurality of weights and a weight to be partially learned among the plurality of weights; and (c) learning the new class by using weights in which the qualitative information has a value less than or equal to the threshold value.
展开▼