18618371. MACHINE-LEARNED MODELS FOR GENERATING CODE SNIPPETS WITH PREDICTED PLACEHOLDERS FOR OPTIMIZING SOFTWARE DEVELOPMENT simplified abstract (GOOGLE LLC)
MACHINE-LEARNED MODELS FOR GENERATING CODE SNIPPETS WITH PREDICTED PLACEHOLDERS FOR OPTIMIZING SOFTWARE DEVELOPMENT
Organization Name
Inventor(s)
Daniel Dun-ning Woo Johnson of Toronto (CA)
Daniel Stefan Tarlow of Montréal (CA)
Maxim Tabachnyk of Munich (DE)
Marc Hatcher Rasi of Sunnyvale CA (US)
Jacob Austin of New York NY (US)
Hassan Abolhassani of Palo Alto CA (US)
Jacob Hanson Hegna of Minneapolis MN (US)
MACHINE-LEARNED MODELS FOR GENERATING CODE SNIPPETS WITH PREDICTED PLACEHOLDERS FOR OPTIMIZING SOFTWARE DEVELOPMENT - A simplified explanation of the abstract
This abstract first appeared for US patent application 18618371 titled 'MACHINE-LEARNED MODELS FOR GENERATING CODE SNIPPETS WITH PREDICTED PLACEHOLDERS FOR OPTIMIZING SOFTWARE DEVELOPMENT
- Simplified Explanation:**
This patent application describes a method for machine-learned code segment prediction to optimize software development. The method involves using a machine-learned code prediction model to complete incomplete segments of code with a high degree of certainty.
- Key Features and Innovation:**
- Obtaining an incomplete segment of code
- Processing the incomplete segment with a machine-learned code prediction model
- Obtaining a set of segment completion predictions
- Determining an aggregated segment completion prediction
- Replacing uncertain portions of the prediction with an input field
- Potential Applications:**
This technology can be applied in software development to assist programmers in completing code segments efficiently and accurately.
- Problems Solved:**
This technology addresses the challenge of predicting and completing code segments accurately, saving time and effort for software developers.
- Benefits:**
- Improved efficiency in software development
- Enhanced accuracy in code completion
- Reduction in errors and debugging time
- Commercial Applications:**
This technology can be utilized in various industries that rely on software development, such as tech companies, IT departments, and software development firms, to streamline the coding process and improve productivity.
- Prior Art:**
Prior art related to this technology may include research on machine learning models for code completion and prediction in software development.
- Frequently Updated Research:**
Researchers may be conducting studies on enhancing machine-learned code prediction models for more accurate and efficient code completion.
- Questions about Machine-Learned Code Segment Prediction:**
1. How does this technology improve the efficiency of software development? 2. What are the potential limitations of using machine-learned code prediction models in completing code segments?
Original Abstract Submitted
Systems and methods of the present disclosure are directed to a method for machine- learned code segment prediction for optimizing software development. The method includes obtaining an incomplete segment of code. The method includes processing the incomplete segment of code with a machine-learned code prediction model to obtain a sampled set of segment completion predictions that include code that completes the incomplete segment of code. The method includes determining an aggregated segment completion prediction from the sampled set of segment completion predictions. The method includes replacing a portion of the aggregated segment completion prediction with an input field, wherein the portion of the aggregated segment completion prediction is associated with a degree of certainty less than a threshold degree of certainty.
