Exploring the Role of Generative AI in Automatic Code Generation for Software Development
Keywords:
Generative AI, Automatic Code Generation, Software Development, Large Language Models, Codex, GitHub CopilotAbstract
Generative Artificial Intelligence (AI) is increasingly revolutionizing software engineering by automating code creation. This paper explores the role of generative AI in automatic code generation, discussing the techniques used, its implications for developer productivity, and the transformative potential it offers in software development workflows. Key technologies such as transformers and pre-trained large language models (LLMs) have enabled tools like OpenAI’s Codex and GitHub Copilot to significantly enhance coding efficiency and reduce errors. Through comprehensive literature review and visual analyses, we examine how this evolution builds upon prior approaches and redefines the programming paradigm
References
Allamanis, M., Barr, E. T., Bird, C., & Sutton, C. (2018). Learning to represent programs with graphs. ICLR. https://openreview.net/pdf?id=BJOFETxR-
Kacheru, G., Bajjuru, R., & Arthan, N. (2019). Security Considerations When Automating Software Development. Revista de Inteligencia Artificial en Medicina, 10(1), 598617.
Hindle, A., Barr, E. T., Gabel, M., Su, Z., & Devanbu, P. (2012). On the naturalness of software. Proceedings of the 34th International Conference on Software Engineering, 837–847. https://doi.org/10.1109/ICSE.2012.6227135
Ahmed, F., Li, L., & Srikumar, V. (2021). InCoder: Code Generation with Cross-Lingual Pretraining. arXiv preprint. https://arxiv.org/abs/2112.10989
Svyatkovskiy, A., Sundaresan, N., Fu, S., & Jiang, N. (2020). IntelliCode Compose: Code generation using transformer. arXiv preprint. https://arxiv.org/abs/2005.08025
Chen, M., Tworek, J., Jun, H., et al. (2021). Evaluating large language models trained on code. arXiv preprint. https://arxiv.org/abs/2107.03374
Kacheru, G. (2021). The Future of Cyber Defence: Predictive Security with Artificial Intelligence. International Journal of Advanced Research in Basic Engineering Sciences and Technology (IJARBEST), 7(12), 46–55.
Vaswani, A., Shazeer, N., Parmar, N., et al. (2017). Attention is all you need. NeurIPS, 5998–6008. https://papers.nips.cc/paper_files/paper/2017/hash/3f5ee243547dee91fbd053c1c4a845aa-Abstract.html
Kacheru, G., Bajjuru, R., & Arthan, N. (2022). Surge of Cyber Scams during the COVID19 Pandemic: Analyzing the Shift in Tactics. BULLET: Jurnal Multidisiplin Ilmu, 1(02), 192202.
Raychev, V., Bielik, P., & Vechev, M. (2016). Probabilistic model for code with decision trees. OOPSLA. https://doi.org/10.1145/2983990.2984041
Hellendoorn, V. J., & Devanbu, P. (2017). Deep learning type inference. Proceedings of the 2017 11th Joint Meeting on Foundations of Software Engineering, 152–162. https://doi.org/10.1145/3106237.3106290
Karampatsis, R. M., & Sutton, C. (2020). Big Code!= Big Vocabulary: Open-Vocabulary Models for Source Code. ICSE. https://doi.org/10.1145/3377811.3380364
Tufano, M., Watson, C., Bavota, G., et al. (2019). Deep learning to detect redundant comments in source code. ICSE, 1110–1121. https://doi.org/10.1109/ICSE.2019.00115
Downloads
Published
Issue
Section
License
Copyright (c) 2023 Dr. Arjun Raj Mehta (Author)
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
