Secure ML Model Generation Platform with Hybrid Cloud Access
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The generation of machine learning models is a complex and time-consuming process, especially when multiple models are required to achieve desired outcomes, and there is a need for secure control over datasets used for model creation, as the owners of the dataset and the models may not be the same.
Innovation Solution
A machine learning model generation platform that allows for the creation and deployment of portable and scalable AI solutions, enabling end-to-end orchestration of AI workflows, efficient packaging of AI pipeline training jobs, and secure access to datasets through a hybrid cloud environment, using a high-level API for training and automated tuning of models, and secure environments for usage tracking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If machine learning models are trained on datasets owned by other entities, then model performance and accuracy can be improved, but data security and access control become problematic
Solution Approach 1:
The patent introduces a secure access intermediary system that mediates between dataset owners and model training processes. This intermediary maintains secure control over the dataset while enabling model training through controlled access mechanisms, allowing model accuracy to improve without compromising data security or ownership rights
2Measurement precision
If multiple machine learning models are generated to achieve desired outcomes, then solution quality and accuracy improve, but the complexity and time required for model generation increase significantly
Solution Approach 1:
The patent creates a universal model generation platform that can handle multiple model training tasks through a single integrated system. This platform provides reusable components and standardized processes that reduce complexity when generating multiple models, allowing the system to maintain high solution accuracy while managing the complexity of producing multiple models efficiently
3Measurement precision
If multiple machine learning models are generated to achieve desired outcomes, then solution quality improves, but the time required for model generation increases
Solution Approach 1:
The patent implements preliminary action by pre-configuring model templates, pre-processing data pipelines, and pre-establishing training frameworks. These preliminary preparations enable faster model generation when multiple models are needed, reducing the time penalty while maintaining the ability to generate high-quality solutions through multiple models
Data Source
AI summary
A method receiving a dataset, storing the dataset in a secured drive, synthesizing a representative dataset, in the secured drive, based on the dataset, granting access to a specialist to view the representative dataset, receiving a model that was generated using the representative dataset, running the model on the representative dataset, validating the results of running the model on the representative dataset, and presenting the validated results of running the model on the representative dataset in a graphical user interface.