METHOD OF EVALUATING ROBUSTNESS OF ARTIFICIAL NEURAL NETWORK WATERMARKING AGAINST MODEL STEALING ATTACKS

    公开(公告)号:US20220164417A1

    公开(公告)日:2022-05-26

    申请号:US17361994

    申请日:2021-06-29

    Abstract: Disclosed is a method of evaluating robustness of artificial neural network watermarking against model stealing attacks. The method of evaluating robustness of artificial neural network watermarking may include the steps of: training an artificial neural network model using training data and additional information for watermarking; collecting new training data for training a copy model of a structure the same as that of the trained artificial neural network model; training the copy model of the same structure by inputting the collected new training data into the copy model; and evaluating robustness of watermarking for the trained artificial neural network model through a model stealing attack executed on the trained copy model.

    NUCLEIC ACID-MEDIATED PATTERN REPLICATION AND METHOD OF MANUFACTURING 2-D MATERIAL USING THE SAME

    公开(公告)号:US20220127662A1

    公开(公告)日:2022-04-28

    申请号:US17510690

    申请日:2021-10-26

    Abstract: Provided are the nucleic acid-mediated pattern replication and a method of manufacturing a 2-D material using the same. A method of manufacturing a 2-D material according to an embodiment may include preparing a first material having a first nucleic acid patterned on a surface thereof, bonding a linker-nucleic acid to the first nucleic acid, bonding the first nucleic acid and a second nucleic acid attached to a surface of a second material through the linker-nucleic acid and replicating a pattern of the first material to the surface of the second material, separating the first material, and applying a third material on a pattern replicated to the surface of the second material.

    METHOD FOR ESTIMATING READ REFERENCE VOLTAGES FOR FLASH STORAGE USING NEURAL NETWORK AND APPARATUS THEREFOR

    公开(公告)号:US20220121387A1

    公开(公告)日:2022-04-21

    申请号:US17076274

    申请日:2020-10-21

    Abstract: Disclosed is a method for estimating a read reference voltage for a flash memory using a neural network and an apparatus therefor. According to an embodiment of the inventive concept, a method for reading a flash memory may include receiving predetermined variable values with respect to a flash memory, estimating a read reference voltage corresponding to the received variable values, using a neural network of a pre-trained learning model with respect to the variable values, and reading the flash memory, using the estimated read reference voltage. The method may further include updating the learning model of the neural network by reflecting a difference in characteristics between individual chips constituting the flash memory.

Patent Agency Ranking