Deep learning (DL) is fundamentally intertwined with software engineering (SE). The application of DL to a computational problem represents a new programming paradigm: rather than developing a program in code, a program is “learned” from large-scale datasets. This can be referred to as SE for DL. On the other hand, DL techniques can be used to automate or improve existing SE tasks. This is referred to as DL for SE. In this case, DL systems can be viewed as an inflection point for software development, as they enable new capabilities that cannot be realized cost-effectively through “traditional” software wherein the behavior of a program must be specified analytically. There currently exists an unprecedented amount of software data that is freely available in open-source software repositories. This data spans several software artifacts, from source code and test code, to requirements and issue tracker data. Given the effectiveness by which DL systems are able to learn representations from such largescale data corpora, there is ample opportunity to leverage DL techniques to help automate or improve a wide range of developer tasks. All these new forms of development carry with a new set of challenges that represent several opportunities for novel research.
Due to the rapid pace at which DL techniques have been adopted, we want to know the research's current successes, failures, and opportunities at the intersection of SE & DL. There is research on identifying that the most extensively investigated sub-processes are software testing and maintenance. In such sub-processes, DL models are widely used to process bug reports, malware classification, libraries, and commits recommendations generation. Some solutions are oriented to effort estimation, classifying software requirements, identifying GUI visual elements, identifying code authors, and the similarity of source codes, predicting and classifying defects, and analyzing bug reports in testing and maintenance processes. However, many open issues still remain to be investigated. How do researchers integrate DL into SE problems? Which SE phases are facilitated by DL? Do practitioners benefit from DL? How datasets are used to develop techniques for particular development tasks, and designing DL models that effectively capture the inherent structure present in a wide range of different software artifacts. The answers help practitioners and researchers develop practical DL models for SE tasks. In an effort to bring clarity to this crosscutting area of work, from its modern inception to the present, the goal of this workshop is to outline high-priority areas for crosscutting research that sits at the intersection of DL and SE.
The topics of interest in the discussion are but not limited to:
We encourage submissions on the topics mentioned above with a page limit of max 8 pages, IEEE format. In addition, we will also allow position papers and short and demonstration papers of two to four pages.
Submitted papers must have been neither previously accepted for publication nor concurrently submitted for review in another journal, book, conference, or workshop.
Full papers (maximum of 8 pages, including references). Original research in the crosscutting area of SE&DL, either empirical, theoretical, or showing the practical experience of DL for SE, or SE for DL.
Short and demonstration papers (maximum of 4 pages, including references). Work that describes novel techniques, tools, ideas, and positions that have yet to be fully developed; or the potential applicability (or not) of the result in an industrial context.
Position papers (maximum of 2 pages, including references). Contributions that analyze trends in SE&DL and raise issues of importance. Position papers are intended to seed discussion and debate at the workshop, and thus will be reviewed with respect to relevance and their ability to spark discussions.
Download IEEE Text Template (.docx)
Submissions are required in PDF format via EasyChair at: https://easychair.org/conferences/?conf=sedl2023.
|Peng Liang||Wuhan university, Wuhan, China|
|Lin Liu||Tsinghua University, Beijing, China|
|Lin Shi||University of Chinese Academy Sciences, Beijing, China|
|Xiaohong Chen||East China Normal University, Shanghai, China|
|Rubing Huang||Macau University of Science and Technology, Macao SAR, China|
|Bo Yang||Beijing Forestry University, Beijing, China|
|Time Schedule||21st March, 2023|
|Conference Location||Orchid Room, Level 28, Hotel Okura|
|9:20-10:10||Ge Li, Peking University. Keynote|
|10:40-11:00||Mingrui Yang, Dalin Zhang, Beijing Jiaotong University. Deep Reinforcement Learning Guided Decision Tree Learning For Program Synthesis|
|11:00-11:20||Rui Zhu, Wenxin Li, Yunnan University. TAG: UML Activity Diagram Deeply Supervised Generation from Business Textural Specification|
|11:20-11:40||Xiangchen Shen, Haibo Chen, Jinfu Chen, Jiawei Zhang, Shuhui Wang, Jiangsu University. EcoDialTest:Adaptive Mutation Schedule for Automated Dialogue Systems Testing|
|11:40-12:00||Chao Zhu, Jing Chen, Rui Zhu, Zhengqiong Wang, Shihan Liu, Jishu Wang, Yunnan University. ASTHGCN: Adaptive Spatio-Temporal Hypergraph Convolutional Network for Traffic Forecasting|
|14:00-14:20||Fengyu Yang, Guangdong Zeng, Fa Zhong, Wei Zheng, Peng Xiao, Nanchang Hangkong University. Interpretable Software Defect Prediction Incorporating Multiple Rules|
|14:20-14:40||Baolei Wang, Xuan Zhang, Kunpeng Du, Chen Gao, Linyu Li, Yunnan university. Multimodal Sentiment Analysis under modality deficiency with prototype-Augmentation in software engineering|
|14:40-15:00||Shifan Liu, Zhanqi Cui, Ruilin Xie, Beijing Information Science and Technology University. SICUP: a Comment Updating Approach based on Structural Information|
|15:00-15:20||Yue Ju, Yixuan Tang, Jinpeng Lan, Xiangbo Mi, Jingxuan Zhang, Nanjing University of Aeronautics and Astronautics . A Cross-Language Name Binding Recognition and Discrimination Approach for Identifiers|