公开(公告)号：US20230342603A1

公开(公告)日：2023-10-26

申请号：US18303976

申请日：2023-04-20

Applicant: Samsung Electronics Co., Ltd.

Inventor： Aditya JHAWAR ,  Rajesh Kumar PANDA ,  Nikhil SAHNI ,  Aarif ZAFAR ,  Renju Chirakarotu NAIR ,  Shubham JAIN ,  Umesh Murlidhar PATIL

IPC: G06N3/08

CPC classification number: G06N3/08

Abstract: A method and an electronic device for secure training of an artificial intelligence (AI) model are provided. The method includes determining, by the electronic device, a first set of layers from a base AI model for performing training in a secure mode, and determining, by the electronic device, a second set of layers other than the first set of layers in the base AI model, simultaneously training, by the electronic device, the first set of layers in the secure mode and the second set of layers in a non-secure mode, and generating, by the electronic device, the AI model by combining the trained first set of layers and the trained second set of layers in the secure mode.

公开(公告)号：US20240152283A1

公开(公告)日：2024-05-09

申请号：US18228354

申请日：2023-07-31

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Aarif ZAFAR ,  Renju Chirakarotu NAIR ,  Nikhil SAHNI ,  Sagar Mahesh MALIYE ,  Aditya JHAWAR ,  Rajesh Kumar PANDA

IPC: G06F3/06 ,  G06N5/04

CPC classification number: G06F3/062 ,  G06F3/0655 ,  G06F3/0673 ,  G06N5/04

Abstract: A method for performing an inference includes: detecting a context among at least one context associated with at least one application; triggering a model execution command to a smart agent of an electronic device, based on the detected context; loading a machine learning (ML) model into a secure storage of the electronic device, based on the detected context and the triggered model execution command; generating, using the loaded ML model, an inference, based on data associated with the detected context; and sharing the generated inference with each application of the at least one application that is registered for the detected context.

公开(公告)号：US20220327222A1

公开(公告)日：2022-10-13

申请号：US17835649

申请日：2022-06-08

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Srividya DESIREDDY ,  Yogesh SHEORAN ,  Sagar Mahesh MALIYE ,  Balwant Singh SHEKHAWAT ,  Aarif ZAFAR ,  Sharmila MANI ,  Rajesh Kumar PANDA ,  Srikanth MANDALAPU ,  Renju Chirakarotu NAIR ,  Nikhil SAHNI

IPC: G06F21/60 ,  G06F21/44 ,  G06N3/10 ,  G06F11/34

Abstract: Embodiments herein provide a method for securing Neural Network (NN) models. The method includes: identifying, by the first electronic device, a crucial layer of a first NN model that is to be deployed for processing in a second electronic device. The method includes extracting, by the first electronic device, the crucial layer of the first NN model. The method includes encrypting, by the first electronic device, the crucial layer. The method includes generating, by the first electronic device, a second NN model includes the encrypted crucial layer. The method includes deploying, by the first electronic device, the second NN model to the second electronic device for processing in the second electronic device.