Client-Server Hybrid AI Scoring for Personalized Actions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In data science applications, existing client-server models often require frequent data transmission from clients to servers for user model management, leading to resource inefficiencies and limitations in modeling user interactions across multiple servers, especially in applications where cross-user learning is important.
Innovation Solution
Implementing a client-server hybrid AI scoring system where clients generate user-specific scores based on historical actions and request server scores using cross-user models, combining both to produce hybrid scores for customized actions, thereby enabling real-time modeling and efficient resource utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If clients transmit data frequently to servers for user model management, then user-specific personalization is improved, but server resource load increases
Solution Approach 1:
The patent segments the AI scoring system into two independent components: a client-side user-specific model that processes local historical actions, and a server-side cross-user model that handles aggregate patterns. This segmentation allows personalization to occur locally without requiring frequent server communication, reducing server load while maintaining personalization accuracy.
Solution Approach 2:
The client performs preliminary processing by generating user-specific scores locally using the downloaded user-specific model before requesting server scores. This preliminary action reduces the need for frequent data transmission and server processing, as the client has already performed initial scoring and filtering locally.
2Adaptability or versatility
If all modeling is done on the server, then cross-user learning is improved, but data transmission requirements increase
Solution Approach 1:
The patent introduces a new dimension to the modeling architecture by implementing client-side modeling capabilities in addition to server-side modeling. This dimensional expansion allows the system to leverage both user-specific local patterns and cross-user aggregate patterns without requiring complete data transmission to the server, as modeling now occurs in both client and server dimensions.
3Measurement precision
If complex data science applications are executed on servers, then model accuracy is improved, but server resource consumption increases
Solution Approach 1:
The patent extracts the user-specific modeling function from the server and places it on the client device. The user-specific model is downloaded to the client, where it independently processes local historical actions to generate personalized scores. This extraction reduces server resource consumption by offloading repetitive user-specific processing while the server maintains cross-user learning capabilities.
Data Source
AI summary
Client-server hybrid A.I. scores for customized actions are described. A client generates client scores corresponding to client customized actions by applying a user-specific model to an action received from a user, the user-specific model based on at least one historical action received from the user. The client requests a server to provide server scores corresponding to server customized actions by applying a cross-user model to the action received from the user, the cross-user model based on historical actions associated with server users. The client generates hybrid scores corresponding to hybrid customized actions by combining the client scores with the server scores, in response to receiving the server scores from the server. The client causes the hybrid customized actions to be outputted based on the corresponding hybrid scores.