DeciX: Explain Deep Learning Based Code Generation Applications
Authors: 
Simin Chen, Zexin Li, 
Wei Yang, Cong LiuAuthors Info & Claims
Simin Chen, Zexin Li, Wei Yang, Cong LiuAuthors Info & ClaimsArticle No.: 107, Pages 2424 - 2446
Abstract
Deep learning-based code generation (DL-CG) applications have shown great potential for assisting developers in programming with human-competitive accuracy. However, lacking transparency in such applications due to the uninterpretable nature of deep learning models makes the automatically generated programs untrustworthy. In this paper, we develop DeciX, a first explanation method dedicated to DL-CG applications. DeciX is motivated by observing two unique properties of DL-CG applications: output-to-output dependencies and irrelevant value and semantic space. These properties violate the fundamental assumptions made in existing explainable DL techniques and thus cause applying existing techniques to DL-CG applications rather pessimistic and even incorrect. DeciX addresses these two limitations by constructing a causal inference dependency graph, containing a novel method leveraging causal inference that can accurately quantify the contribution of each dependency edge in the graph to the end prediction result. Proved by extensive experiments assessing popular, widely-used DL-CG applications and several baseline methods, DeciX is able to achieve significantly better performance compared to state-of-the-art in terms of several critical performance metrics, including correctness, succinctness, stability, and overhead. Furthermore, DeciX can be applied to practical scenarios since it does not require any knowledge of the DL-CG model under explanation. We have also conducted case studies that demonstrate the applicability of DeciX in practice.
References
[1]
Julius Adebayo, Justin Gilmer, Michael Muelly, Ian Goodfellow, Moritz Hardt, and Been Kim. 2018. Sanity checks for saliency maps. Advances in neural information processing systems, 31 (2018).
[2]
Daniel Arp, Michael Spreitzenbarth, Malte Hubner, Hugo Gascon, Konrad Rieck, and CERT Siemens. 2014. Drebin: Effective and explainable detection of android malware in your pocket. In Ndss. 14, 23–26.
[3]
Yoshua Bengio, Réjean Ducharme, and Pascal Vincent. 2000. A neural probabilistic language model. Advances in neural information processing systems, 13 (2000).
[4]
Hila Chefer, Shir Gur, and Lior Wolf. 2021. Transformer interpretability beyond attention visualization. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 782–791.
[5]
Simin Chen, Soroush Bateni, Sampath Grandhi, Xiaodi Li, Cong Liu, and Wei Yang. 2020. DENAS: automated rule generation by knowledge extraction from neural networks. In Proceedings of the 28th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Software Engineering. 813–825.
[6]
Simin Chen, Xiaoning Feng, Xiaohong Han, Cong Liu, and Wei Yang. 2024. PPM: Automated Generation of Diverse Programming Problems for Benchmarking Code Generation Models. arXiv preprint arXiv:2401.15545.
[7]
Simin Chen, Hamed Khanpour, Cong Liu, and Wei Yang. 2022. Learn to Reverse DNNs from AI Programs Automatically. In Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence, IJCAI 2022, Vienna, Austria, 23-29 July 2022, Luc De Raedt (Ed.). ijcai.org, 666–672. https://doi.org/10.24963/IJCAI.2022/94
[8]
Jürgen Cito, Isil Dillig, Seohyun Kim, Vijayaraghavan Murali, and Satish Chandra. 2021. Explaining mispredictions of machine learning models using rule induction. In Proceedings of the 29th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Software Engineering. 716–727.
[9]
Jürgen Cito, Isil Dillig, Vijayaraghavan Murali, and Satish Chandra. 2022. Counterfactual Explanations for Models of Code. In 2022 IEEE/ACM 44th International Conference on Software Engineering: Software Engineering in Practice (ICSE-SEIP). 125–134.
[10]
Jay DeYoung, Sarthak Jain, Nazneen Fatema Rajani, Eric Lehman, Caiming Xiong, Richard Socher, and Byron C Wallace. 2019. ERASER: A benchmark to evaluate rationalized NLP models. arXiv preprint arXiv:1911.03429.
[11]
EMNLP. 2022. Most Influential EMNLP Papers (2021-02). https://www.paperdigest.org/2021/02/most-influential-emnlp-papers/
[12]
Zhangyin Feng, Daya Guo, Duyu Tang, Nan Duan, Xiaocheng Feng, Ming Gong, Linjun Shou, Bing Qin, Ting Liu, and Daxin Jiang. 2020. CodeBERT: A pre-trained model for programming and natural languages. arXiv preprint arXiv:2002.08155.
[13]
Ruth C Fong and Andrea Vedaldi. 2017. Interpretable explanations of black boxes by meaningful perturbation. In Proceedings of the IEEE international conference on computer vision. 3429–3437.
[14]
Daniel Fried, Armen Aghajanyan, Jessy Lin, Sida Wang, Eric Wallace, Freda Shi, Ruiqi Zhong, Wen-tau Yih, Luke Zettlemoyer, and Mike Lewis. 2022. Incoder: A generative model for code infilling and synthesis. arXiv preprint arXiv:2204.05999.
[15]
Timon Gehr, Matthew Mirman, Dana Drachsler-Cohen, Petar Tsankov, Swarat Chaudhuri, and Martin Vechev. 2018. Ai2: Safety and robustness certification of neural networks with abstract interpretation. In 2018 IEEE Symposium on Security and Privacy (SP). 3–18.
[16]
Xiaodong Gu, Hongyu Zhang, and Sunghun Kim. 2018. Deep code search. In 2018 IEEE/ACM 40th International Conference on Software Engineering (ICSE). 933–944.
[17]
Xiaodong Gu, Hongyu Zhang, Dongmei Zhang, and Sunghun Kim. 2016. Deep API learning. In Proceedings of the 2016 24th ACM SIGSOFT international symposium on foundations of software engineering. 631–642.
[18]
Wenbo Guo, Dongliang Mu, Jun Xu, Purui Su, Gang Wang, and Xinyu Xing. 2018. Lemna: Explaining deep learning based security applications. In proceedings of the 2018 ACM SIGSAC conference on computer and communications security. 364–379.
[19]
Yaru Hao, Li Dong, Furu Wei, and Ke Xu. 2021. Self-attention attribution: Interpreting information interactions inside transformer. In Proceedings of the AAAI Conference on Artificial Intelligence. 35, 12963–12971.
[20]
Yaru Hao, Li Dong, Furu Wei, and Ke Xu. 2021. Self-attention attribution: Interpreting information interactions inside transformer. In Proceedings of the AAAI Conference on Artificial Intelligence. 35, 12963–12971.
[21]
Xincheng He, Lei Xu, Xiangyu Zhang, Rui Hao, Yang Feng, and Baowen Xu. 2021. PyART: Python API recommendation in real-time. In 2021 IEEE/ACM 43rd International Conference on Software Engineering (ICSE). 1634–1645.
[22]
HuggingFace. 2022. HuggingFace NMT Helsinki-NLP/opus-mt-de-en. https://huggingface.co
[23]
Gareth James, Daniela Witten, Trevor Hastie, and Robert Tibshirani. 2013. An introduction to statistical learning. 112, Springer.
[24]
Mintong Kang, Nezihe Merve Gürel, Ning Yu, Dawn Song, and Bo Li. 2024. C-RAG: Certified Generation Risks for Retrieval-Augmented Language Models. arXiv preprint arXiv:2402.03181.
[25]
Svetoslav Karaivanov, Veselin Raychev, and Martin Vechev. 2014. Phrase-based statistical translation of programming languages. In Proceedings of the 2014 ACM International Symposium on New Ideas, New Paradigms, and Reflections on Programming & Software. 173–184.
[26]
Been Kim, Rajiv Khanna, and Oluwasanmi O Koyejo. 2016. Examples are not enough, learn to criticize! criticism for interpretability. Advances in neural information processing systems, 29 (2016).
[27]
Seohyun Kim, Jinman Zhao, Yuchi Tian, and Satish Chandra. 2021. Code prediction by feeding trees to transformers. In 2021 IEEE/ACM 43rd International Conference on Software Engineering (ICSE). 150–162.
[28]
Goro Kobayashi, Tatsuki Kuribayashi, Sho Yokoi, and Kentaro Inui. 2020. Attention is Not Only a Weight: Analyzing Transformers with Vector Norms. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP). 7057–7075.
[29]
Himabindu Lakkaraju, Stephen H Bach, and Jure Leskovec. 2016. Interpretable decision sets: A joint framework for description and prediction. In Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining. 1675–1684.
[30]
Jiwei Li, Will Monroe, and Dan Jurafsky. 2016. Understanding neural networks through representation erasure. arXiv preprint arXiv:1612.08220.
[31]
Qi Liu, Matt J Kusner, and Phil Blunsom. 2020. A survey on contextual embeddings. arXiv preprint arXiv:2003.07278.
[32]
Yue Liu, Chakkrit Tantithamthavorn, Li Li, and Yepang Liu. 2022. Explainable AI for Android Malware Detection: Towards Understanding Why the Models Perform So Well? In the 33rd International Symposium on Software Engineering Reliability.
[33]
Scott M Lundberg and Su-In Lee. 2017. A unified approach to interpreting model predictions. Advances in neural information processing systems, 30 (2017).
[34]
Scott M Lundberg and Su-In Lee. 2017. A Unified Approach to Interpreting Model Predictions. In Advances in Neural Information Processing Systems 30, I. Guyon, U. V. Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, and R. Garnett (Eds.). Curran Associates, Inc., 4765–4774. http://papers.nips.cc/paper/7062-a-unified-approach-to-interpreting-model-predictions.pdf
[35]
MEGVII. 2021. Online Face Verification. https://docs.microsoft.com/en-us/visualstudio/intellicode/intellicode-visual-studio
[36]
MicroSoft. 2021. CodeXGlue. https://github.com/microsoft/CodeXGLUE/
[37]
MicroSoft. 2022. Copilot. https://copilot.github.com/
[38]
Tomas Mikolov, Kai Chen, Greg Corrado, and Jeffrey Dean. 2013. Efficient estimation of word representations in vector space. arXiv preprint arXiv:1301.3781.
[39]
Tim Miller. 2019. Explanation in artificial intelligence: Insights from the social sciences. Artificial intelligence, 267 (2019), 1–38.
[40]
Anh Tuan Nguyen, Michael Hilton, Mihai Codoban, Hoan Anh Nguyen, Lily Mast, Eli Rademacher, Tien N Nguyen, and Danny Dig. 2016. API code recommendation using statistical learning from fine-grained changes. In Proceedings of the 2016 24th ACM SIGSOFT International Symposium on Foundations of Software Engineering. 511–522.
[41]
Trong Duc Nguyen, Anh Tuan Nguyen, and Tien N Nguyen. 2016. Mapping API elements for code migration with vector representations. In 2016 IEEE/ACM 38th International Conference on Software Engineering Companion (ICSE-C). 756–758.
[42]
Trong Duc Nguyen, Anh Tuan Nguyen, and Tien N Nguyen. 2016. Mapping API elements for code migration with vector representations. In 2016 IEEE/ACM 38th International Conference on Software Engineering Companion (ICSE-C). 756–758.
[43]
Trong Duc Nguyen, Anh Tuan Nguyen, Hung Dang Phan, and Tien N Nguyen. 2017. Exploring API embedding for API usages and applications. In 2017 IEEE/ACM 39th International Conference on Software Engineering (ICSE). 438–449.
[44]
Weili Nie, Yang Zhang, and Ankit Patel. 2018. A theoretical explanation for perplexing behaviors of backpropagation-based visualizations. In International Conference on Machine Learning. 3809–3818.
[45]
Erik Nijkamp, Bo Pang, Hiroaki Hayashi, Lifu Tu, Huan Wang, Yingbo Zhou, Silvio Savarese, and Caiming Xiong. 2022. Codegen: An open large language model for code with multi-turn program synthesis. arXiv preprint arXiv:2203.13474.
[46]
Hammond Pearce, Baleegh Ahmad, Benjamin Tan, Brendan Dolan-Gavitt, and Ramesh Karri. 2021. An Empirical Cybersecurity Evaluation of GitHub Copilot’s Code Contributions. arXiv preprint arXiv:2108.09293.
[47]
Kexin Pei, Yinzhi Cao, Junfeng Yang, and Suman Jana. 2017. Deepxplore: Automated whitebox testing of deep learning systems. In proceedings of the 26th Symposium on Operating Systems Principles. 1–18.
[48]
Sebastian Proksch, Johannes Lerch, and Mira Mezini. 2015. Intelligent code completion with Bayesian networks. ACM Transactions on Software Engineering and Methodology (TOSEM), 25, 1 (2015), 1–31.
[49]
PyGPT2. 2021. Automatic complete python codes. https://huggingface.co/SIC98/GPT2-python-code-generator
[50]
Yao Qiang, Deng Pan, Chengyin Li, Xin Li, Rhongho Jang, and Dongxiao Zhu. 2022. Attcat: Explaining transformers via attentive class activation tokens. Advances in Neural Information Processing Systems, 35 (2022), 5052–5064.
[51]
Veselin Raychev, Martin Vechev, and Eran Yahav. 2014. Code completion with statistical language models. In Proceedings of the 35th ACM SIGPLAN Conference on Programming Language Design and Implementation. 419–428.
[52]
Marco Tulio Ribeiro, Sameer Singh, and Carlos Guestrin. 2016. " Why should i trust you?" Explaining the predictions of any classifier. In Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining. 1135–1144.
[53]
Roei Schuster and Congzheng Song. 2021. You autocomplete me: Poisoning vulnerabilities in neural code completion. In USENIX Security.
[54]
Ramprasaath R Selvaraju, Michael Cogswell, Abhishek Das, Ramakrishna Vedantam, Devi Parikh, and Dhruv Batra. 2017. Grad-cam: Visual explanations from deep networks via gradient-based localization. In Proceedings of the IEEE international conference on computer vision. 618–626.
[55]
Avanti Shrikumar, Peyton Greenside, and Anshul Kundaje. 2017. Learning important features through propagating activation differences. In International conference on machine learning. 3145–3153.
[56]
Karen Simonyan, Andrea Vedaldi, and Andrew Zisserman. 2013. Deep inside convolutional networks: Visualising image classification models and saliency maps. arXiv preprint arXiv:1312.6034.
[57]
Karen Simonyan, Andrea Vedaldi, and Andrew Zisserman. 2014. Visualising image classification models and saliency maps. Deep Inside Convolutional Networks.
[58]
Jost Tobias Springenberg, Alexey Dosovitskiy, Thomas Brox, and Martin Riedmiller. 2014. Striving for simplicity: The all convolutional net. arXiv preprint arXiv:1412.6806.
[59]
Alexey Svyatkovskiy, Ying Zhao, Shengyu Fu, and Neel Sundaresan. 2019. Pythia: AI-assisted code completion system. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 2727–2735.
[60]
Yuchi Tian, Kexin Pei, Suman Jana, and Baishakhi Ray. 2018. Deeptest: Automated testing of deep-neural-network-driven autonomous cars. In Proceedings of the 40th international conference on software engineering. 303–314.
[61]
Yao Wan, Wei Zhao, Hongyu Zhang, Yulei Sui, Guandong Xu, and Hai Jin. 2022. What do they capture? a structural analysis of pre-trained language models for source code. In Proceedings of the 44th International Conference on Software Engineering. 2377–2388.
[62]
Yu Wang, Ke Wang, and Linzhang Wang. 2021. WheaCha: A method for explaining the predictions of models of code. arXiv preprint arXiv:2102.04625.
[63]
Yue Wang, Weishi Wang, Shafiq Joty, and Steven CH Hoi. 2021. Codet5: Identifier-aware unified pre-trained encoder-decoder models for code understanding and generation. arXiv preprint arXiv:2109.00859.
[64]
Xin Xin, Jinlong Li, and Zeqi Tan. 2021. N-ary constituent tree parsing with recursive semi-Markov model. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2631–2642.
[65]
Zhou Yang, Jieke Shi, Junda He, and David Lo. 2022. Natural attack for pre-trained models of code. In Proceedings of the 44th International Conference on Software Engineering. 1482–1493.
[66]
Matthew D Zeiler and Rob Fergus. 2014. Visualizing and understanding convolutional networks. In European conference on computer vision. 818–833.
[67]
Zhengran Zeng, Hanzhuo Tan, Haotian Zhang, Jing Li, Yuqun Zhang, and Lingming Zhang. 2022. An extensive study on pre-trained models for program understanding and generation. In Proceedings of the 31st ACM SIGSOFT international symposium on software testing and analysis. 39–51.
[68]
Xinze Zhang, Junzhe Zhang, Zhenhua Chen, and Kun He. 2021. Crafting Adversarial Examples for Neural Machine Translation. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). Association for Computational Linguistics, Online. 1967–1977. https://doi.org/10.18653/v1/2021.acl-long.153
[69]
Zhaowei Zhang, Hongyu Zhang, Beijun Shen, and Xiaodong Gu. 2022. Diet code is healthy: Simplifying programs for pre-trained models of code. In Proceedings of the 30th ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering. 1073–1084.
[70]
Luisa M Zintgraf, Taco S Cohen, Tameem Adel, and Max Welling. 2017. Visualizing deep neural network decisions: Prediction difference analysis. arXiv preprint arXiv:1702.04595.
Index Terms
- DeciX: Explain Deep Learning Based Code Generation Applications
Recommendations
Explaining Explanations: An Empirical Study of Explanations in Code Reviews
Code reviews are central for software quality assurance. Ideally, reviewers should explain their feedback to enable authors of code changes to understand the feedback and act accordingly. Different developers might need different explanations in different ...
Embedding API dependency graph for neural code generation
AbstractThe problem of code generation from textual program descriptions has long been viewed as a grand challenge in software engineering. In recent years, many deep learning based approaches have been proposed, which can generate a sequence of code from ...
Explainability Metrics and Properties for Counterfactual Explanation Methods
Explainable and Transparent AI and Multi-Agent SystemsAbstractThe increasing application of Explainable AI (XAI) methods to enhance the transparency and trustworthiness of AI systems designates the need to quantitatively assess and analyze the theoretical and behavioral characteristics of explanations ...
Comments
Information & Contributors
Information
Published In
July 2024
2770 pages
Copyright © 2024 Owner/Author.
This work is licensed under a Creative Commons Attribution International 4.0 License.
Publisher
Association for Computing Machinery
New York, NY, United States
Publication History
Published: 12 July 2024
Published in PACMSE Volume 1, Issue FSE
Author Tags
Qualifiers
- Research-article
Funding Sources
- CCF
- CNS
- CPS
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 414Total Downloads
- Downloads (Last 12 months)414
- Downloads (Last 6 weeks)94
Reflects downloads up to 27 Dec 2024
Other Metrics
Citations
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in