公开(公告)号：GB2606867A

公开(公告)日：2022-11-23

申请号：GB202207563

申请日：2020-11-05

Applicant: IBM

Inventor： JAYARAM KALLAPALAYAM RADHAKRISHNAN ,  GEGI THOMAS ,  ASHISH VERMA

IPC: H04L9/00

Abstract: Embodiments relate to training a machine learning model based on an iterative algorithm in a distributed, federated, private, and secure manner. Participating entities are registered in a collaborative relationship. The registered participating entities are arranged in a topology and a topological communication direction is established. Each registered participating entity receives a public additive homomorphic encryption (AHE) key and local machine learning model weights are encrypted with the received public key. The encrypted local machine learning model weights are selectively aggregated and distributed to one or more participating entities in the topology responsive to the topological communication direction. The aggregated sum of the encrypted local machine learning model weights is subjected to decryption with a corresponding private AHE key. The decrypted aggregated sum of the encrypted local machine learning model weights is shared with the registered participating entities.

公开(公告)号：GB2529363B

公开(公告)日：2016-06-01

申请号：GB201521768

申请日：2014-05-22

Applicant: IBM

Inventor： JOSEPH WILLIAM LIGMAN ,  MARCO PISTOIA ,  GEGI THOMAS ,  OMER TRIPP

IPC: G06F9/50 ,  H04L29/08 ,  H04W52/02

Abstract: A mobile device includes a computer-readable medium storing computer program instructions, a data processor to execute the instructions, and communication circuitry configured for local area wireless connectivity with neighboring mobile devices and for wireless connectivity to a remote server from which at least a portion of a data set is downloaded. Execution of the computer program instructions results in estimating a cost to perform a computation task on the data set. If the estimated cost is greater than a threshold cost, an ad-hoc wireless network is formed with at least one other mobile device and the mobile device downloads a portion of the data set assigned to the mobile device. The mobile device then performs a computation task on the downloaded portion of the data set and wirelessly transfers a result of the computation task to the at least one other mobile device of the ad-hoc wireless network.