18024889. METHOD FOR PERFORMING FEDERATED LEARNING IN WIRELESS COMMUNICATION SYSTEM, AND APPARATUS THEREFOR simplified abstract (LG ELECTRONICS INC.)
Contents
METHOD FOR PERFORMING FEDERATED LEARNING IN WIRELESS COMMUNICATION SYSTEM, AND APPARATUS THEREFOR
Organization Name
Inventor(s)
METHOD FOR PERFORMING FEDERATED LEARNING IN WIRELESS COMMUNICATION SYSTEM, AND APPARATUS THEREFOR - A simplified explanation of the abstract
This abstract first appeared for US patent application 18024889 titled 'METHOD FOR PERFORMING FEDERATED LEARNING IN WIRELESS COMMUNICATION SYSTEM, AND APPARATUS THEREFOR
Simplified Explanation
The abstract describes a method for base stations in a wireless communication system to perform federated learning. Federated learning is a machine learning approach where multiple devices collaborate to train a shared model without sharing their raw data.
- The base station receives local parameters from multiple terminals.
- It obtains an integrated parameter by combining the local parameters.
- The integrated parameter is then transmitted back to each terminal.
- The method removes biases in the local parameters using the collective information.
Potential applications of this technology:
- Artificial intelligence: The base station and terminals can be linked to an AI module, allowing for collaborative learning and improved AI capabilities.
- Unmanned aerial vehicles (UAVs): UAVs can benefit from federated learning to enhance their navigation and decision-making abilities.
- Robotics: Federated learning can be used to train robots collectively, enabling them to learn from each other's experiences and improve their performance.
- Augmented reality (AR) and virtual reality (VR): The technology can be applied to AR and VR devices to enhance user experiences and provide personalized content.
- 6G services: The base station and terminals can be utilized in the development of future 6G services, leveraging federated learning for improved network performance and user satisfaction.
Problems solved by this technology:
- Privacy concerns: By performing federated learning, the raw data from individual terminals is not shared, addressing privacy issues associated with centralized data collection.
- Communication efficiency: Instead of transmitting large amounts of data to a central server, the integrated parameter is sent back to each terminal, reducing communication overhead.
- Bias removal: The method removes biases in the local parameters, ensuring fair and accurate learning across all terminals.
Benefits of this technology:
- Improved privacy: Users can participate in collaborative learning without compromising the privacy of their data.
- Enhanced learning performance: By combining local parameters, the integrated parameter can provide a more comprehensive and accurate representation of the collective knowledge.
- Reduced communication costs: Transmitting the integrated parameter instead of raw data reduces the amount of data that needs to be transmitted, leading to more efficient communication.
- Scalability: The method can be applied to a large number of terminals, allowing for scalable federated learning in wireless communication systems.
Original Abstract Submitted
Disclosed is a method by which a base station performs federated learning in a wireless communication system. A method, according to one embodiment of the present disclosure, comprises: receiving, by a base station, a plurality of local parameters from a plurality of terminals; obtaining an integrated parameter on the basis of the plurality of local parameters; and transmitting the integrated parameter to each of the plurality of terminals, wherein at least one bias included in the plurality of local parameters is removed by using the plurality of local parameters. The base station/terminals of the present disclosure may be linked to an artificial intelligence, module, an unmanned aerial vehicle (UAV), a robot, an augmented reality (AR) device, a virtual reality (VR) device, a device related to a 6G service, and the like.
