AsiaInfo Technologies has filed a patent for a federated learning system in a telecom communication system. The system involves an intelligent service provider acting as a master and coordinating with additional intelligent service providers as slaves to perform federated learning. The service-related information obtained from the slave providers is merged with the master provider’s information to provide intelligent services to the requester or provide merged information to the slave providers. GlobalData’s report on AsiaInfo Technologies gives a 360-degree view of the company including its patenting strategy. Buy the report here.
Smarter leaders trust GlobalData
According to GlobalData’s company profile on AsiaInfo Technologies, digital lending was a key innovation area identified from patents. AsiaInfo Technologies's grant share as of June 2023 was 1%. Grant share is based on the ratio of number of grants to total number of patents.
Federated learning in a telecom system with intelligent service providers
A telecom communication system patent application (Publication Number: US20230146099A1) describes an intelligent service provider that utilizes federated learning to provide intelligent services. The intelligent service provider acts as a master intelligent service provider and coordinates with one or more additional intelligent service providers acting as slave intelligent service providers.
The intelligent service provider receives service requests and performs federated learning in coordination with the slave intelligent service providers. It merges service-related information obtained from the slave intelligent service providers with its own service-related information to provide intelligent services for the service requester. Alternatively, it can provide the merged service-related information to the slave intelligent service providers.
The intelligent service provider can serve as the master intelligent service provider based on several principles. These principles include being the first to receive the service request, being coordinated as the master by multiple service requesters, being pre-configured as the master, or having better service capability than the slave intelligent service providers.
The service capability of the intelligent service provider is determined by factors such as computing capability and resource utilization rate. The involvement of the slave intelligent service providers in federated learning is determined based on the availability of trained data and performance indicators provided by the additional intelligent service providers. Performance indicators can include throughput, reliability, and delay.
The processing circuitry of the intelligent service provider can switch federated learning on or off based on a request from the service requester or the performance of the intelligent service provider. It can also determine the availability of service-related information provided by the slave intelligent service provider based on the activity level or performance-related information of the slave intelligent service provider.
The service-related information can include artificial intelligence models, model parameters, or partial service results. The intelligent service provider can merge the models or model parameters provided by the slave intelligent service providers to build a global intelligent service model. It can also merge the partial service results to provide an intelligent service.
The intelligent service provider can provide feedback to the slave intelligent service providers for updating the global intelligent model parameters. The slave intelligent service providers can provide model parameters to the master intelligent service provider periodically or based on a request from the master.
In summary, the patent application describes an intelligent service provider in a telecom communication system that utilizes federated learning and coordination with slave intelligent service providers to provide intelligent services. The intelligent service provider merges service-related information obtained from the slave intelligent service providers and can switch federated learning on or off based on various factors. It can also provide feedback and update model parameters for improved performance.
To know more about GlobalData’s detailed insights on AsiaInfo Technologies, buy the report here.
Data Insights
From
The gold standard of business intelligence.
Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.
GlobalData, the leading provider of industry intelligence, provided the underlying data, research, and analysis used to produce this article.
GlobalData Patent Analytics tracks bibliographic data, legal events data, point in time patent ownerships, and backward and forward citations from global patenting offices. Textual analysis and official patent classifications are used to group patents into key thematic areas and link them to specific companies across the world’s largest industries.
