Mixed Reality Source Code Debugging via ML Output Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a need to protect source code and its underlying logic from viewers while allowing users with security clearance to view and fix bugs, and to continuously generate output data based on source code without revealing the logic, even in real-time, in a secure, efficient, and accurate manner.
Innovation Solution
A system that trains a source code output machine learning model using source code statements, input data, and expected output data, allowing it to generate sample output data for comparison, and uses a mixed reality apparatus to securely render the output to users, enabling changes to the source code without exposing the underlying logic.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If source code is made visible to users for debugging and viewing, then users can identify and fix bugs, but source code security and logic protection are compromised
Solution Approach 1:
The patent creates a virtual copy of the source code environment within the mixed reality system. The ML model generates output that replicates the functionality of the source code without exposing the actual code. Users interact with this virtual representation to debug and view logic, while the original source code remains protected in the secure environment.
2Measurement precision
If source code logic is exposed to generate output data in real-time, then output generation is accurate and transparent, but security clearance requirements are violated
Solution Approach 1:
The patent introduces a machine learning model as an intermediary between the protected source code and the output generation process. The ML model is trained on the source code logic and then independently generates output data, serving as a mediator that preserves accuracy while maintaining security. The mixed reality apparatus further mediates the presentation of this output to authorized users.
3Reliability
If manual processing of source code is used to generate output, then security can be maintained, but processing speed and efficiency are reduced
Solution Approach 1:
The patent performs preliminary training of the machine learning model on the source code logic before deployment. This preliminary action captures the essential processing rules and patterns, enabling the model to rapidly generate output without requiring real-time manual analysis of the source code. The security protection is maintained because the trained model operates independently after the initial training phase.
4Speed
If source code statements are continuously processed to generate output, then real-time updates are achieved, but computational resources are consumed
Solution Approach 1:
The patent replaces the computationally intensive process of continuously analyzing source code statements with a pre-trained machine learning model that has copied the essential processing logic. The model generates output rapidly without requiring continuous access to or processing of the original source code, significantly reducing computational resource consumption while maintaining real-time capability.
Data Source
AI summary
Systems, computer program products, and methods are described herein for implementing a mixed reality apparatus to automatically and dynamically generate output based on source code. The present disclosure is configured to identify at least one source code statement; train the source code output machine learning model with the at least one source code statement; generate at least one input data for the at least one source code statement, the at least one input data is associated with an at least one expected output data; apply the at least one input data to the source code output machine learning model; output the at least one sample output data; transmit the at least one sample output data to a mixed reality apparatus; and compare the at least one sample output data and the at least one expected output data and determine whether the at least one source code statement is valid.