InferCode: Self-Supervised Learning of Code Representations by Predicting Subtrees

• Main Authors: Bui, Nghi D. Q., Yu, Yijun, Jiang, Lingxiao
• Format: Conference Proceeding
• Language: English
• Subjects: Cloning; code clone detection; code retrieval; code search; Computer bugs; Computing methodologies; Computing methodologies > Machine learning; Computing methodologies > Machine learning > Machine learning approaches; cross language; fine tuning; Predictive models; self supervised; Software and its engineering; Software and its engineering > Software creation and management; Software and its engineering > Software creation and management > Software development techniques; Software and its engineering > Software creation and management > Software post-development issues; Software and its engineering > Software notations and tools; Software and its engineering > Software notations and tools > General programming languages; Software engineering; Syntactics; Task analysis; Training; unlabel data; unlabelled data
• ISSN: 1558-1225; 1558-1225
• DOI: 10.1109/ICSE43902.2021.00109

Learning code representations has found many uses in software engineering, such as code classification, code search, comment generation, and bug prediction, etc. Although representations of code in tokens, syntax trees, dependency graphs, paths in trees, or the combinations of their variants have been proposed, existing learning techniques have a major limitation that these models are often trained on datasets labeled for specific downstream tasks, and as such the code representations may not be suitable for other tasks. Even though some techniques generate representations from unlabeled code, they are far from being satisfactory when applied to the downstream tasks. To overcome the limitation, this paper proposes InferCode, which adapts the self-supervised learning idea from natural language processing to the abstract syntax trees (ASTs) of code. The novelty lies in the training of code representations by predicting subtrees automatically identified from the contexts of ASTs. With InferCode, subtrees in ASTs are treated as the labels for training the code representations without any human labelling effort or the overhead of expensive graph construction, and the trained representations are no longer tied to any specific downstream tasks or code units. We have trained an instance of InferCode model using Tree-Based Convolutional Neural Network (TBCNN) as the encoder of a large set of Java code. This pre-trained model can then be applied to downstream unsupervised tasks such as code clustering, code clone detection, cross-language code search, or be reused under a transfer learning scheme to continue training the model weights for supervised tasks such as code classification and method name prediction. Compared to prior techniques applied to the same downstream tasks, such as code2vec, code2seq, ASTNN, using our pre-trained InferCode model higher performance is achieved with a significant margin for most of the tasks, including those involving different programming languages. The implementation of InferCode and the trained embeddings are available at the link: https://github.com/bdqnghi/infercode.