Google Inc.

X Development LLC

Qianyu Zhang

Bin Ni

Olivia Hatalsky

Rishabh Singh

Georgios Evangelopoulos

Owen Lewis

Zhiqiang Yuan

Grigory Bronevetsky

Catalina Codruta Cangea

Benoit Schillings

Implementations are described herein for adapting existing source code snippets to new contexts. In various implementations, a command may be detected to incorporate an existing source code snippet into destination source code. An embedding may be generated based on the existing source code snippet, e.g., by processing the existing source code snippet using an encoder. The destination source code may be processed to identify one or more decoder constraints. Subject to the one or more decoder constraints, the embedding may be processed using a decoder to generate a new version of the existing source code snippet that is adapted to the destination source code.

Adapting existing source code snippets to new contexts

Techniques are described herein for training a machine learning model and using the trained machine learning model to more accurately determine alignments between matching/corresponding nodes of predecessor and successor graphs representing predecessor and successor source code snippets. A method includes: obtaining a first abstract syntax tree that represents a predecessor source code snippet and a second abstract syntax tree that represents a successor source code snippet; determining a mapping across the first and second abstract syntax trees; obtaining a first control-flow graph that represents the predecessor source code snippet and a second control-flow graph that represents the successor source code snippet; aligning blocks in the first control-flow graph with blocks in the second control-flow graph; and applying the aligned blocks as inputs across a trained machine learning model to generate an alignment of nodes in the first abstract syntax tree with nodes in the second abstract syntax tree.

Matching graphs generated from source code

Implementations are described herein for leveraging prior source code transformations to facilitate automatic creation and/or recommendation of tools for automating aspects of source code transformations captured in real time. In various implementations, a transformation made by a programmer to a source code snipped may be captured in a source code editor application in real time. Based on the transformation and the intent, one or more candidate source code transformations may be identified from one or more repositories of prior source code transformations made by one or more other programmers. The source code editor application may be caused to provide output indicative of a tool that is operable to automate one or more edits associated with both the transformation made by the programmer to the source code snippet and with one or more of the candidate source code transformations.

Generation and/or recommendation of tools for automating aspects of computer programming

Techniques are described herein for translating source code in one programming language to source code in another programming language using machine learning. In various implementations, one or more components of one or more generative adversarial networks, such as a generator machine learning model, may be trained to generate 'synthetically-naturalistic' source code that can be used as a translation of source code in an unfamiliar language. In some implementations, a discriminator machine learning model may be employed to aid in training the generator machine learning model, e.g., by being trained to discriminate between human-generated ('genuine') and machine-generated ('synthetic') source code.

Translating between programming languages using machine learning

Techniques are described herein for using artificial intelligence to 'learn,' statistically, a target programming style that is imposed in and/or evidenced by a code base. Once the target programming style is learned, it can be used for various purposes. In various implementations, one or more generative adversarial networks ('GANs'), each including a generator machine learning model and a discriminator machine learning model, may be trained to facilitate learning and application of target programming style(s). In some implementations, the discriminator(s) and/or generator(s) may operate on graphical input, and may take the form of graph neural networks (&#x201c;GNNs&#x201d;), graph attention neural networks (&#x201c;GANNs&#x201d;), graph convolutional networks (&#x201c;GCNs&#x201d;), etc., although this is not required.

Learning and using programming styles

Implementations are described herein for training and using machine learning to determine mappings between matching nodes of graphs representing predecessor source code snippets and graphs representing successor source code snippets. In various implementations, first and second graphs may be obtained, wherein the first graph represents a predecessor source code snippet and the second graph represents a successor source code snippet. The first graph and the second graph may be applied as inputs across a trained machine learning model to generate node similarity measures between individual nodes of the first graph and nodes of the second graph. Based on the node similarity measures, a mapping may be determined across the first and second graphs between pairs of matching nodes.

Growing community of inventors

Sunnyvale, CA, United States of America

Qianyu Zhang