TY - GEN
T1 - MulCode
T2 - 28th IEEE International Conference on Software Analysis, Evolution and Reengineering, SANER 2021
AU - Wang, Deze
AU - Yu, Yue
AU - Li, Shanshan
AU - Dong, Wei
AU - Wang, Ji
AU - Qing, Liao
N1 - Publisher Copyright:
© 2021 IEEE.
PY - 2021/3
Y1 - 2021/3
N2 - Recent years have witnessed the significant rise of Deep Learning (DL) techniques applied to source code. Researchers exploit DL for a multitude of tasks and achieve impressive results. However, most tasks are explored separately, resulting in a lack of generalization of the solutions. In this work, we propose MulCode, a multi-task learning approach for source code understanding that learns unified representation space for tasks, with the pre-trained BERT model for the token sequence and the Tree-LSTM model for abstract syntax trees. Furthermore, we integrate two source code views into a hybrid representation via the attention mechanism and set learnable uncertainty parameters to adjust the tasks' relationship.We train and evaluate MulCode in three downstream tasks: comment classification, author attribution, and duplicate function detection. In all tasks, MulCode outperforms the state-of-the-art techniques. Moreover, experiments on three unseen tasks demonstrate the generalization ability of MulCode compared with state-of-the-art embedding methods.
AB - Recent years have witnessed the significant rise of Deep Learning (DL) techniques applied to source code. Researchers exploit DL for a multitude of tasks and achieve impressive results. However, most tasks are explored separately, resulting in a lack of generalization of the solutions. In this work, we propose MulCode, a multi-task learning approach for source code understanding that learns unified representation space for tasks, with the pre-trained BERT model for the token sequence and the Tree-LSTM model for abstract syntax trees. Furthermore, we integrate two source code views into a hybrid representation via the attention mechanism and set learnable uncertainty parameters to adjust the tasks' relationship.We train and evaluate MulCode in three downstream tasks: comment classification, author attribution, and duplicate function detection. In all tasks, MulCode outperforms the state-of-the-art techniques. Moreover, experiments on three unseen tasks demonstrate the generalization ability of MulCode compared with state-of-the-art embedding methods.
KW - attention mechanism
KW - deep learning
KW - multi-task learning
KW - representation learning
UR - https://www.scopus.com/pages/publications/85106562192
U2 - 10.1109/SANER50967.2021.00014
DO - 10.1109/SANER50967.2021.00014
M3 - 会议稿件
AN - SCOPUS:85106562192
T3 - Proceedings - 2021 IEEE International Conference on Software Analysis, Evolution and Reengineering, SANER 2021
SP - 48
EP - 59
BT - Proceedings - 2021 IEEE International Conference on Software Analysis, Evolution and Reengineering, SANER 2021
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 9 March 2021 through 12 March 2021
ER -