Improving Automatic Source Code Summarization via Deep Reinforcement Learning

Code summarization provides a high level natural language description of the function performed by code, as it can benefit the software maintenance, code categorization and retrieval. To the best of our knowledge, most state-of-the-art approaches follow an encoder-decoder framework which encodes the...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2018 33rd IEEE/ACM International Conference on Automated Software Engineering (ASE) s. 397 - 407
Hlavní autoři: Wan, Yao, Zhao, Zhou, Yang, Min, Xu, Guandong, Ying, Haochao, Wu, Jian, Yu, Philip S.
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: ACM 01.09.2018
Témata:
Code summarization
Codes
comment generation
Decoding
deep learning
Deep reinforcement learning
Long short term memory
Measurement
Natural languages
reinforcement learning
Software maintenance
Source coding
Syntactics
Vectors
ISSN:2643-1572
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Code summarization provides a high level natural language description of the function performed by code, as it can benefit the software maintenance, code categorization and retrieval. To the best of our knowledge, most state-of-the-art approaches follow an encoder-decoder framework which encodes the code into a hidden space and then decode it into natural language space, suffering from two major drawbacks: a) Their encoders only consider the sequential content of code, ignoring the tree structure which is also critical for the task of code summarization; b) Their decoders are typically trained to predict the next word by maximizing the likelihood of next ground-truth word with previous ground-truth word given. However, it is expected to generate the entire sequence from scratch at test time. This discrepancy can cause an exposure bias issue, making the learnt decoder suboptimal. In this paper, we incorporate an abstract syntax tree structure as well as sequential content of code snippets into a deep reinforcement learning framework (i.e., actor-critic network). The actor network provides the confidence of predicting the next word according to current state. On the other hand, the critic network evaluates the reward value of all possible extensions of the current state and can provide global guidance for explorations. We employ an advantage reward composed of BLEU metric to train both networks. Comprehensive experiments on a real-world dataset show the effectiveness of our proposed model when compared with some state-of-the-art methods.
ISSN:2643-1572
DOI:10.1145/3238147.3238206