US Patent Application 17714884. GRADIENT GROUPING FOR COMPRESSION IN FEDERATED LEARNING FOR MACHINE LEARNING MODELS simplified abstract

GRADIENT GROUPING FOR COMPRESSION IN FEDERATED LEARNING FOR MACHINE LEARNING MODELS

Inventors

Eren BALEVI of San Diego CA (US)

Taesang YOO of San Diego CA (US)

GRADIENT GROUPING FOR COMPRESSION IN FEDERATED LEARNING FOR MACHINE LEARNING MODELS - A simplified explanation of the abstract

• This abstract for appeared for patent application number 17714884 Titled 'GRADIENT GROUPING FOR COMPRESSION IN FEDERATED LEARNING FOR MACHINE LEARNING MODELS'

Simplified Explanation

This abstract describes a method of wireless communication using a machine learning model for federated learning. The user equipment (UE) receives the machine learning model from a network entity and computes a set of gradient vector parameters using a local dataset. These parameters are then grouped into multiple subsets. The method further computes representative values for each subset by averaging the gradients within each subset. Finally, the representative values are transmitted back to the network entity for the first communication round of the federated learning process.

Original Abstract Submitted

A method of wireless communication, by a user equipment (UE), includes receiving, from a network entity, a machine learning model for federated learning. The method also includes computing a set of gradient vector parameters during a first communication round of the federated learning for the machine learning model using a local dataset. The method further includes grouping the set of gradient vector parameters of the machine learning model into multiple subsets. The method also includes computing a representative value of all gradients within each of the subsets to obtain representative values for each of the subsets. The method includes transmitting the representative values to the network entity for the first communication round of the federated learning.