US Patent Application 17828582. RECONCILING COMPUTING INFRASTRUCTURE AND DATA IN FEDERATED LEARNING simplified abstract

From WikiPatents
Jump to navigation Jump to search

RECONCILING COMPUTING INFRASTRUCTURE AND DATA IN FEDERATED LEARNING

Organization Name

Cisco Technology, Inc.

Inventor(s)

Myungjin Lee of Bellevue WA (US)

Gaoxiang Luo of Minneapolis MN (US)

Ramana Rao V. R. Kompella of Cupertino CA (US)

RECONCILING COMPUTING INFRASTRUCTURE AND DATA IN FEDERATED LEARNING - A simplified explanation of the abstract

This abstract first appeared for US patent application 17828582 titled 'RECONCILING COMPUTING INFRASTRUCTURE AND DATA IN FEDERATED LEARNING

Simplified Explanation

The abstract describes a controller for a federated learning system that organizes the computing infrastructure as a tree structure. The controller establishes connections between datasets and nodes in the tree structure. It receives instructions to train models in the federated learning system using specific datasets and configures the system accordingly.

  • The patent application describes a controller for a federated learning system.
  • The controller organizes the computing infrastructure of the system as a tree structure.
  • It establishes associations between datasets and nodes in the tree structure.
  • The controller receives instructions to train models in the system using specific datasets.
  • It configures the system to perform the model training using the specified datasets and the tree structure.


Original Abstract Submitted

In one embodiment, a controller for a federated learning system represents computing infrastructure for the federated learning system as a tree structure. The controller forms associations between datasets available to the federated learning system and nodes in the tree structure. The controller receives one or more instructions to perform model training in the federated learning system with datasets specified using their associations. The controller configures, in response to the one or more instructions, the federated learning system to perform the model training using the datasets specified by the one or more instructions using the tree structure.