AI Model Deployment for Rule Engine Updates

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Updating rule engines using artificial intelligence models is hindered by the need for specific training data and the complexity of interpreting rule engine intricacies, which is resource-intensive and labor-intensive.

Innovation Solution

The system detects when a rule engine should be updated and generates code to effectively use current model components without requiring specific training data, by using artificial intelligence models to modify inputs and generate outputs for comparison, and by training generative models to improve rules-based decision engines.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Difficulty of detecting and measuring

If artificial intelligence models are trained to directly create or update rule engines, then the ability to interpret rule engine complexities is improved, but the resource intensity and labor intensity for training data preparation increases significantly

Engineering Contradiction:

Improveability to interpret rule engine complexitiesVSAvoidtime for training data preparation

Core Design Contradiction:

Difficulty of detecting and measuringVSLoss of time

Solution Approach 1:

The patent introduces an intermediary evaluation system that assesses rule engine updates without requiring extensive training data. Instead of training AI models directly on rule engine complexities, the system uses an intermediate evaluation layer that measures update effectiveness through predefined criteria and metrics, thereby reducing the time and resources needed for training data preparation while maintaining the ability to interpret rule engine complexities

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent implements preliminary evaluation frameworks and metrics that are established before actual rule engine updates occur. By pre-defining evaluation criteria, measurement protocols, and assessment methodologies, the system prepares the groundwork for interpreting rule engine complexities without requiring extensive training data preparation at the time of update, thus reducing time loss

Inventive Principle:

Principle #10Preliminary action

2Measurement precision

If training data is made specific to the model being updated and the objective being sought, then the quality of training for effective model learning is improved, but the data gathering and generation process becomes more resource-intensive and labor-intensive

Engineering Contradiction:

Improvequality of training for model learningVSAvoidvolumes of training data required

Core Design Contradiction:

Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent develops universal evaluation frameworks and measurement criteria that can be applied across different rule engine updates and AI models without requiring custom training data for each specific case. These multi-functional evaluation tools assess model learning effectiveness and update quality across various contexts, thereby maintaining high training quality while reducing the volumes of specialized training data required

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent uses synthetic training data and simulated rule engine environments that replicate real-world scenarios without requiring extensive actual training data. By creating copied or simulated versions of training scenarios, the system maintains high training quality for model learning while significantly reducing the resource-intensive data gathering and generation processes

Inventive Principle:

Principle #26Copying

3Measurement precision

If the system compares outputs of modified inputs with actual inputs to detect update requirements, then the detection accuracy of rule engine update needs is improved, but the computational resources and processing time increase

Engineering Contradiction:

Improvedetection accuracy of update requirementsVSAvoidcomputational resources for output comparison

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent implements a tiered comparison approach that performs partial output comparisons based on the specific context and risk level of rule engine updates. Instead of always performing exhaustive output comparisons between modified and actual inputs, the system selectively applies comparison depth based on update criticality, thereby maintaining high detection accuracy for critical updates while reducing computational resource consumption for routine updates

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS12210858B1Systems and methods for detecting required rule engine updated using artificial intelligence models

Publication Date: 2025.01.28 CITIBANK N A

AI summary

The systems and methods provide a model deployment criterion. The model deployment criterion indicates a difference in a value against which the proxy model may be measured to determine when, if ever, the proxy model should be deployed to replace the existing rule engine. The model deployment criterion may be keyed to the proxy model (e.g., based on a difference in its size, throughput speed, number of changes, etc.), the existing rule engine (e.g., based on a difference in its age, update occurrences to its rule base, etc.), and/or comparisons between models (e.g., based on differences in results, throughput speed, efficiency, etc.).