ML Model Discovery Architecture for Scalable Reuse
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing machine learning models are typically created ad-hoc for each application, lacking scalability and reusability, and current architectures do not facilitate easy expansion or reuse without significant modification.
Innovation Solution
A computing architecture that includes a machine learning-based discovery model to identify and recommend machine learning models from a repository, utilizing an aggregator layer to combine outputs from multiple models, and provides a graphical interface for model management and activation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If machine learning models are created ad-hoc for each application, then each application can have customized functionality, but the models cannot be reused and the architecture is not scalable
Solution Approach 1:
The patent creates a universal model repository that stores machine learning models that can be reused across multiple applications. The system allows a single model to serve multiple purposes and applications through the discovery model that identifies suitable models based on application requirements, eliminating the need to create separate ad-hoc models for each application while maintaining customized functionality through model selection and configuration.
2Productivity
If a centralized model repository is created to enable reuse, then model reusability improves, but the complexity of managing and discovering models increases
Solution Approach 1:
The patent introduces a discovery model as an intermediary component that sits between the model repository and applications. This discovery model simplifies the complexity by automatically identifying and recommending suitable models from the repository based on application requirements, eliminating the need for applications to directly manage or search the repository, thus reducing architectural complexity while maintaining high reusability.
3Measurement precision
If developers manually search and select models from a repository, then model selection can be precise, but the process is time-consuming and not automated
Solution Approach 1:
The patent implements a self-service system where the discovery model automatically identifies and recommends suitable machine learning models from the repository based on application requirements and parameters. This automation eliminates the need for developers to manually search and evaluate models, reducing selection time significantly while maintaining high accuracy through the intelligent matching capabilities of the discovery model that considers various model characteristics and application needs.
Data Source
AI summary
A computing architecture is optimized for the reuse of machine learning model. A request is received for a machine learning model recommendation that specifies parameters associated with a desired machine learning model (e.g., schema, etc.). Thereafter, a machine learning-based discovery model, recommends at least one machine learning model (to reuse) based on the parameters specified in the request. Next, data characterizing the recommended at least one machine learning model is provided (e.g., loaded into memory, displayed in a graphical user interface, transmitted to a remote computing system, and/or stored in physical persistence, etc.). In some variations, a graphical user interface can be rendered that allows a client application to select one of the recommended machine learning models and, further optionally, to activate the selected machine learning models. Related apparatus, systems, techniques and articles are also described.