Search

Project-Based Teaching in a Deep Learning Course: Practice in a Data Science Program at an Application-Oriented University

Author(s)

Junli Zhang1,*, Zimeng Wang2

Affiliation(s)

1Xi'an Eurasia University, Xi'an, Shaanxi, China 2Electronic Information Engineering, Xi'an University of Posts and Telecommunications, Xi'an, Shaanxi, China *Corresponding Author

Abstract

Deep learning courses are conceptually abstract and involve a long engineering chain. Organized as lectures plus verification labs, they often leave students in an application-oriented data science program able to run code without being able to explain why a model works, and unable to transfer what they have practised to a new problem. This paper reports a project-based teaching practice in a 48-hour Deep Learning course. The content was reorganized into seven modules; instruction followed a path of problem-driven learning, knowledge construction and transfer; project tasks were arranged at four levels, from reproducing classic models to transfer, real application and final presentation; and five module assignments, one unit test and one project report carried the process assessment. A nine-step class procedure gave students about one third of the contact hours for self-study, group inquiry and coding. Over several rounds, project work was completed more often, topics moved from given datasets to real ones such as sheep breed recognition and leaf disease segmentation, and instructors spotted misunderstandings earlier by asking students to explain code on the spot. Limited contact hours, unclear individual contribution in group work and the assessment difficulty introduced by generative AI tools are also discussed.

Keywords

Deep Learning Course; Project-Based Teaching; Application-Oriented University; Process Assessment; Teaching Practice

References

[1] Zhong Q B, Shao Q J, Xie X H, et al. Construction of a project-based teaching model for engineering in local application-oriented universities: a case study of the Robot Institute of Ningbo University of Technology. Research in Higher Education of Engineering, 2025(3): 61-66+78. (in Chinese) [2] Liu X F, Chen W, Han X, et al. Challenges and opportunities of experimental computer teaching in the era of large language models. Experimental Technology and Management, 2025, 42(8): 36-44. (in Chinese) [3] Lavado-Anguera S, Velasco-Quintana P J, Terrón-López M J. Project-based learning (PBL) as an experiential pedagogical methodology in engineering education: a review of the literature. Education Sciences, 2024, 14(6): 617. [4] Blumenfeld P C, Soloway E, Marx R W, et al. Motivating project-based learning: sustaining the doing, supporting the learning. Educational Psychologist, 1991, 26(3/4): 369-398. [5] Liu J F, Zhong Z X. Research on the project-based learning (PBL) model. Studies in Foreign Education, 2002, 29(11): 18-22. (in Chinese) [6] Prince M. Does active learning work? A review of the research. Journal of Engineering Education, 2004, 93(3): 223-231. [7] Ma Y K, Li X W, Gu L S, et al. Exploration of a research-oriented teaching model for the deep learning course. Journal of Higher Education, 2026(6): 129-132, 137. (in Chinese) [8] Zhang J L. Exploration and practice of a machine learning course based on OBE and moral education. Journal of China Multimedia and Network Teaching, 2024(4). (in Chinese) [9] Szegedy C, Liu W, Jia Y, et al. Going deeper with convolutions Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2015: 1-9. [10]He K, Zhang X, Ren S, et al. Deep residual learning for image recognition Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2016: 770-778. [11]Wu D, Wu H J. Potential risks of the educational application of general large models and how to avoid them: from the perspective of technological ethics. Journal of East China Normal University (Educational Sciences), 2024, 42(8): 64-75. (in Chinese) [12]Ling J, Zhang L L, Chen H G, et al. Application of generative artificial intelligence in experimental computer teaching. Computer Education, 2025(5): 28-32. (in Chinese).