Runtime AI Model Update via Process Reinitialization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current AI/ML model update processes require system downtime and scheduled maintenance, impacting productivity and service availability.
Innovation Solution
A computer-implemented method that listens for update requests for AI/ML models during runtime, reinitializing or re-instantiating the digital process to call an updated version of the AI/ML model, allowing for dynamic updates without downtime.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If AI/ML model update process is implemented using traditional methods, then model updates can be performed, but system downtime and scheduled maintenance are required, impacting productivity and service availability
Solution Approach 1:
The system dynamically switches between old and new model versions during runtime. The digital process is reinitialized to call the updated model version without requiring system shutdown, enabling adaptive model updates while maintaining continuous service availability and productivity
Solution Approach 2:
The updated AI/ML model is prepared and loaded into memory before it is needed for production. The system listens for update requests and reinitializes the digital process to call the pre-loaded updated model, ensuring seamless transitions without downtime
2Ease of manufacture
If AI/ML model updates require system downtime, then model updates can be performed, but service availability and revenue generation are impacted
Solution Approach 1:
The system maintains continuous service operation during model updates. The digital process is reinitialized to call the updated model version while the system remains operational, ensuring uninterrupted service availability and continuous revenue generation
Solution Approach 2:
The system uses an intermediary mechanism (model version management and digital process reinitialization) to transition between old and new model versions. This intermediary layer enables smooth model updates without requiring system shutdown, maintaining service availability
3Adaptability or versatility
If traditional model update methods are used, then updates can be performed, but scheduled maintenance time is required, impacting productivity
Solution Approach 1:
The system dynamically adapts to model update requests by reinitializing the digital process to call the updated model version. This dynamic approach eliminates rigid scheduled maintenance requirements, allowing updates to occur flexibly without impacting productivity
Solution Approach 2:
The system automatically listens for update requests and performs self-updating by reinitializing the digital process to call the updated model version. This self-service capability eliminates the need for manual scheduled maintenance, maintaining full productivity during updates
Data Source
AI summary
Dynamically updating, or retraining and updating, artificial intelligence (AI)/machine learning (ML) models in digital processes at runtime is disclosed. Production operation may not need to be stopped for AI/ML model update or retraining and update. The update steps and/or retraining steps for the AI/ML model may be included as part of the digital process. The AI/ML model update may be requested from internal logic (e.g., from the evaluation of a condition, by an expression that calls for the AI/ML model, etc.), external requests (e.g., from external triggers in a finite state machine (FSM), such as a file change, database data, a service call, etc.), or both. Automation of AI/ML model updates or retraining and updates may be provided, where the software reloads/reinitializes/re-instantiates with a retrained and/or updated AI/ML model after (and possibly immediately after) the AI/ML model becomes available.