SELF-ADAPTIVE NEURAL NETWORKS

专利名称：SELF-ADAPTIVE NEURAL NETWORKS发明人：TAN, Xuan,NEDOVIC, Nikola申请号：EP17150148.9申请日：20170103公开号：EP3276541A1公开日：20180131

专利附图：

摘要：A method of updating a neural network may be provided. A method mayinclude computing gradients for an operating matrix of a current layer of the neuralnetwork based on data of at least one of the current layer and at least other layer of theneural network. The method may also include updating the operating matrix based on

the computed gradients. Further, the method may include updating an indexing matrix ofthe current layer based on the updated operating matrix.

申请人：FUJITSU LIMITED

地址：1-1, Kamikodanaka 4-chome, Nakahara-ku Kawasaki-shi Kanagawa 211-8588 JP

国籍：JP

代理机构：Schultes, Stephan

更多信息请下载全文后查看