-
公开(公告)号：US11100399B2
公开(公告)日：2021-08-24
申请号：US15818877
申请日：2017-11-21
Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION
Inventor： Wei Shan Dong , Peng Gao , Chang Sheng Li , Chun Yang Ma , Kai AD Yang , Ren Jie Yao , Ting Yuan , Jun Zhu
Abstract: Systems and methods for training a neural network model are disclosed. In the method, training data is obtained by a deep neural network (DNN) first, the deep neural network comprising at least one hidden layer. Then features of the training data are obtained from a specified hidden layer of the at least one hidden layer, the specified hidden layer being connected respectively to a supervised classification network for classification tasks and an autoencoder based reconstruction network for reconstruction tasks. And at last the DNN, the supervised classification network and the reconstruction network are trained as a whole based on the obtained features, the training being guided by the classification tasks and the reconstruction tasks.
-
公开(公告)号：US20190156211A1
公开(公告)日：2019-05-23
申请号：US15818877
申请日：2017-11-21
Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION
Inventor： Wei Shan Dong , Peng Gao , Chang Sheng Li , Chun Yang Ma , Kai AD Yang , Ren Jie Yao , Ting Yuan , Jun Zhu
Abstract: Systems and methods training a model are disclosed. In the method, training data is obtained by a deep neural network (DNN) first, the deep neural network comprising at least one hidden layer. Then features of the training data are obtained from a specified hidden layer of the at least one hidden layer, the specified hidden layer being connected respectively to a supervised classification network for classification tasks and an autoencoder based reconstruction network for reconstruction tasks. And at last the DNN, the supervised classification network and the reconstruction network are trained as a whole based on the obtained features, the training being guided by the classification tasks and the reconstruction tasks
-
Search Results
2
records
(took 0.014 seconds)
