专利检索 ap:("Deeping Source Inc.") AND inv:"JEONG, Jong Hu" 第 1 页

1.

发明公开
METHOD FOR GENERATING OBFUSCATED IMAGE TO BE USED IN TRAINING LEARNING NETWORK AND LABELING DEVICE USING THE SAME 审中-实审

公开(公告)号：EP4409470A1

公开(公告)日：2024-08-07

申请号：EP22899127.9

申请日：2022-11-28

申请人： Deeping Source Inc.

发明人： JEONG, Jong Hu

IPC分类号： G06N5/02 , G06N20/00 , G06F16/51 , G06F16/55 , G06V10/774 , G06F21/62 , G06N3/08

CPC分类号： G06F21/6254 , G06N3/084 , G06V20/52 , G06V10/82

2.

发明公开
LEARNING METHOD AND LEARNING DEVICE FOR TRAINING OBFUSCATION NETWORK CAPABLE OF OBFUSCATING ORIGINAL DATA FOR PRIVACY AND TESTING METHOD AND TESTING DEVICE USING THE SAME 审中-公开

公开(公告)号：EP4441694A1

公开(公告)日：2024-10-09

申请号：EP22899126.1

申请日：2022-11-28

申请人： Deeping Source Inc.

发明人： JEONG, Jong Hu

IPC分类号： G06T3/00 , G06F21/62 , G06N20/00

CPC分类号： G06F21/6245 , G06N3/047 , G06N3/048 , G06N3/08 , G06N20/00 , G06N3/04 , G06N3/088 , G06N3/084 , G06N3/045

3.

发明公开
LEARNING METHOD AND LEARNING DEVICE FOR TRAINING OBFUSCATION NETWORK CAPABLE OF OBFUSCATING ORIGINAL DATA FOR PRIVACY TO ACHIEVE INFORMATION RESTRICTION OBFUSCATION AND TESTING METHOD AND TESTING DEVICE USING THE SAME 审中-公开

公开(公告)号：EP4266216A1

公开(公告)日：2023-10-25

申请号：EP22169013.4

申请日：2022-04-20

申请人： Deeping Source Inc.

发明人： JEONG, Jong Hu

IPC分类号： G06N3/04 , G06N3/08

摘要： A learning method for training an obfuscation network, including steps of: (a) inputting a training data into the obfuscation network to (i) extract features and thus generate a data representation by performing a learning operation on the training data and (ii) transform the data representation and thus generate an anonymized data representation, and (b) inputting the anonymized data representation into a task learning network to (i) perform a task by using the anonymized data representation and thus output a task result, (ii) generate a task loss by referring to the task result and its corresponding ground truth, (iii) train the task learning network through a first backpropagation of the task loss such that the task loss is minimized, and (iv) train the obfuscation network through a second backpropagation of the task loss such that the task loss is minimized.