18062158. STOCHASTIC LAYER-WISE AVERAGING AGGREGATION FOR FEDERATED LEARNING simplified abstract (Dell Products L.P.)
Contents
STOCHASTIC LAYER-WISE AVERAGING AGGREGATION FOR FEDERATED LEARNING
Organization Name
Inventor(s)
Isabella Costa Maia of São Paulo (BR)
Iam Palatnik De Sousa of Rio de Janeiro (BR)
Maira Beatriz Hernandez Moran of Rio de Janeiro (BR)
Paulo Abelha Ferreira of Rio de Janeiro (BR)
Pablo Nascimento Da Silva of Niterói (BR)
STOCHASTIC LAYER-WISE AVERAGING AGGREGATION FOR FEDERATED LEARNING - A simplified explanation of the abstract
This abstract first appeared for US patent application 18062158 titled 'STOCHASTIC LAYER-WISE AVERAGING AGGREGATION FOR FEDERATED LEARNING
The abstract of the patent application describes a method where a central node selects a subset of edge nodes from a group that forms a federation. The central node then queries the subset of edge nodes for updates to a global model it maintains. The central node receives updates from the edge nodes and updates the global model accordingly.
- Stochastic selection of a subset of edge nodes by a central node
- Querying the subset of edge nodes for updates to a global model
- Receiving updates from the edge nodes and updating the global model
- Collective definition of a federation by the group of edge nodes
Potential Applications: - Distributed computing systems - Internet of Things (IoT) networks - Collaborative machine learning models
Problems Solved: - Efficient updating of global models in federated systems - Minimizing communication overhead between central and edge nodes
Benefits: - Improved scalability in distributed systems - Enhanced privacy and security in federated learning environments
Commercial Applications: Title: "Optimized Federated Learning System for IoT Networks" This technology could be used in industries such as healthcare, finance, and manufacturing to improve data processing and model training in IoT networks.
Prior Art: There are existing methods for federated learning in distributed systems, but this approach focuses on optimizing the selection and updating process of edge nodes.
Frequently Updated Research: Ongoing research in federated learning systems aims to enhance communication protocols and privacy-preserving techniques for edge computing environments.
Questions about Federated Learning Systems: 1. How does this method improve the efficiency of updating global models in federated systems? 2. What are the key challenges in implementing this technology in real-world IoT networks?
Original Abstract Submitted
One method includes stochastically selecting, by a central node, a subset of edge nodes from a group of edge nodes that collectively defines a federation, querying, by the central node, the edge nodes of the subset for updates to a global model maintained by the central node, receiving, by the central node from the edge nodes of the subset, respective updates to one or more layers of the global model, and updating, by the central node, the global model, using the updates received from the edge nodes of the subset.
