METHOD OF CONTROLLING MACHINE LEARNING MODEL AND SYSTEM PERFORMING THE SAME

    公开(公告)号:US20230237374A1

    公开(公告)日:2023-07-27

    申请号:US17950562

    申请日:2022-09-22

    CPC classification number: G06N20/00 H04L9/0825

    Abstract: A method is provided. The method includes: generating a public key and a secret key, based on a master key and a parameter vector indicating parameters of a first layer of a machine learning model; generating encrypted data by encrypting, based on the public key, an input vector indicating input data of the machine learning model; generating decrypted data corresponding to an approximation value of an inner product of the parameter vector and the input vector by decrypting the encrypted data based on the secret key; and controlling the machine learning model based on the decrypted data to obtain an output value.

    MEMORY CONTROLLER CALCULATING OPTIMAL READ LEVEL, MEMORY SYSTEM INCLUDING THE SAME, AND OPERATING METHOD OF MEMORY CONTROLLER

    公开(公告)号:US20230057932A1

    公开(公告)日:2023-02-23

    申请号:US17685024

    申请日:2022-03-02

    Abstract: Provided are a memory controller calculating an optimal read level, a memory system including the memory controller, and an operating method of the memory controller. The memory controller includes: a processor configured to control a memory operation on the memory device; and a read level calculation module configured to: receive N counting values corresponding to N read levels generated based on a counting operation on data read by using a plurality of read levels, model at least two cell count functions having selected read levels that are selected from the N read levels as inputs, and the N counting values corresponding to the selected read levels as outputs, and calculate an optimal read level based on an optimal cell count function selected from the at least two cell count functions, wherein N is an integer equal to or greater than four, wherein the N counting values include counting values corresponding to at least four different read levels.

    FUNCTIONAL ENCRYPTION SYSTEM AND METHOD OF PERFORMING FUNCTIONAL ENCRYPTION

    公开(公告)号:US20230299955A1

    公开(公告)日:2023-09-21

    申请号:US18096687

    申请日:2023-01-13

    CPC classification number: H04L9/0866 H04L9/14 H04L9/3093

    Abstract: A functional encryption system includes a key generation device configured to generate a public key and a secret key, based on a master key and a parameter vector, an encryption device configured to generate encrypted data by encrypting an input vector, based on the public key, and a decryption device configured to generate decrypted data corresponding to an approximation value of an inner product of the parameter vector and the input vector by decrypting the encrypted data based on the secret key. Security of data used in machine learning, biometric authentication, etc. may be reinforced and attacks using quantum computers may be blocked by incurring random noise in the decrypted data based on the functional encryption using the LWE problem and the lattice problem.

    HOMOMORPHIC ENCRYPTION PROCESSING DEVICE, SYSTEM INCLUDING THE SAME AND METHOD OF PERFORMING HOMOMORPHIC ENCRYPTION PROCESSING

    公开(公告)号:US20210344479A1

    公开(公告)日:2021-11-04

    申请号:US17115161

    申请日:2020-12-08

    Abstract: A homomorphic encryption processing device includes the processing circuitry is configured to generate ciphertext operation level information based on field information. The field information represents a technology field to which homomorphic encryption processing is applied. The ciphertext operation level information represents a maximum number of multiplication operations between homomorphic ciphertexts without a bootstrapping process. The processing circuitry is further configured to select and output a homomorphic encryption parameter based on the ciphertext operation level information. The processing circuitry is further configured to perform one of a homomorphic encryption, a homomorphic decryption and a homomorphic operation, based on the homomorphic encryption parameter. The homomorphic encryption processing device may adaptively generate a homomorphic encryption parameter according to a ciphertext operation level information determined based on a field information, and may perform a homomorphic encryption, a homomorphic decryption and a homomorphic operation based on the homomorphic encryption parameter.

Patent Agency Ranking