Representing source code in vector space to detect errors

US11334467B2 · kind B2 · utility

Assignee

• International Business Machines Corporation · US

Inventors

• David Wehr · Sigourney, US
• Eleanor Pence · Cambridge, US
• Halley Fede · Cambridge, US
• Isabella Yamin · San Francisco, US
• Alexander Sobran · Chapel Hill, US
• Bo Zhang · Cary, US

Key dates

• Filing date: May 3, 2019
• Grant date: May 17, 2022
• Expiry date: Feb 4, 2041

Classification

• Technology area (CPC G): Physics
• CPC primary: G06N3/0895
• WIPO field: Computer technology
• WIPO sector: Electrical engineering

Abstract

A computer-implemented method, system and computer program product for representing source code in vector space. The source code is parsed into an abstract syntax tree, which is then traversed to produce a sequence of tokens. Token embeddings may then be constructed for a subset of the sequence of tokens, which are inputted into an encoder artificial neural network (“encoder”) for encoding the token embeddings. A decoder artificial neural network (“decoder”) is initialized with a final internal cell state of the encoder. The decoder is run the same number of steps as the encoding performed by the encoder. After running the decoder and completing the training of the decoder to learn the inputted token embeddings, the final internal cell state of the encoder is used as the code representation vector which may be used to detect errors in the source code.