PERSONALIZED LEARNING PATHWAYS FOR STUDENTS USING ADAPTIVE AI AND LIMITED EDUCATIONAL DATA
DOI:
https://doi.org/10.63503/c.acset.2025.5Keywords:
Personalized Learning, Adaptive AI, Limited Data, Data Sparsity, Meta-Learning, Transfer Learning, Student Modeling, Educational Data MiningAbstract
Adaptive systems based on Artificial Intelligence (AI) are highly dependent on the paradigm of personalized education, which focuses on tailoring learning experiences to the individual needs of students. The major obstacle to effective and widespread adoption of these systems is that they require high volumes of student interaction data. This reliance poses a critical issue in data-sparse settings, as with new students (the cold-start problem), on niche or advanced courses with limited enrolment, or where data privacy rules are highly restrictive. As a result, the possibility of real personalization is not realized in many cases of practical education. To overcome this inherent setback, a new Hybrid Meta-Transfer Learning (HMTL) framework is presented. HMTL framework is designed to create effective customized learning pathways with minimal initial data. It does so by a synergistic, two-stage architecture: a Transfer Learning component, which is used to construct a robust, generalized Foundational Knowledge Core with data-rich source domains, and a Meta-Learning-based Rapid Adaptation Engine, which is then used to specialize this Core to a new student using only a small number of initial interactions. An extensive simulation shows that the suggested framework outperforms baseline models by a very large margin in making predictions about student performance as well as in terms of the efficiency of the produced learning pathways. This paper introduces the HMTL framework with its architecture, mathematical formulation, and empirical validation, showing a major step forward in the development of a really personalized learning experience using minimal educational data.
References
1. Song, H., Musabirov, I., Bhattacharjee, A., Durand, A., Franklin, M., Rafferty, A., & Williams, J. J. (2025). Adaptive Experiments Under Data Sparse Settings: Applications for Educational Platforms. arXiv preprint arXiv:2501.03999.
2. Lan, A. S., & Baraniuk, R. G. (2017, March). A contextual-bandit approach to personalized learning action selection. In 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 2432-2436). IEEE.
3. Li, L., Chu, W., Langford, J., & Schapire, R. E. (2010). A contextual-bandit approach to personalized news article recommendation. In Proceedings of the 19th international conference on World wide web (pp. 661-670).
4. Shi, C., Shen, C., & Yang, J. (2021). Federated multi-armed bandits with personalization. In Proceedings of the 24th International Conference on Artificial Intelligence and Statistics (pp. 2917-2925). PMLR.
5. Pan, S. J., & Yang, Q. (2009). A survey on transfer learning. IEEE Transactions on knowledge and data engineering, 22(10), 1345-1359.
6. Finn, C., Abbeel, P., & Levine, S. (2017, July). Model-agnostic meta-learning for fast adaptation of deep networks. In International conference on machine learning (pp. 1126-1135). PMLR.
7. Corbett, A. T., & Anderson, J. R. (1994). Knowledge tracing: A model of student learning. User Modeling and User-Adapted Interaction, 4(4), 253-278.
8. Piech, C., Spencer, J., Huang, J., Ganguli, S., Sahami, M., Guibas, L.: Deep knowledge tracing. arXiv preprint arXiv:1506.05908 (2015). https://arxiv.org/abs/1506.05908
9. Antoniou, A., Storkey, A., & Edwards, H. (2017). Data augmentation generative adversarial networks. arXiv preprint arXiv:1711.04340.
10. Arora, S. (2025). Transforming AI Decision Support System with Knowledge Graphs & CAG. International Journal on Engineering Artificial Intelligence Management, Decision Support, and Policies, 2(2), 15-23.
11. Foglino, F., Christiano, P., & Abbeel, P. (2019). Unsupervised Curricula for Visual Meta-Reinforcement Learning. In Advances in Neural Information Processing Systems, 32.
12. Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., & Bengio, Y. (2014). Generative adversarial nets. In Advances in neural information processing systems, 27.
13. Jiang, Y., Luan, S., & Wang, W. Y. (2019). Hierarchical automatic curriculum learning for sparse reward navigation tasks. Neurocomputing, 365, 316-326.
14. Klink, P., D'Eramo, C., Peters, J., & Pajarinen, J. (2022, June). Constrained optimal transport for curriculum reinforcement learning. In International Conference on Machine Learning (pp. 11501-11516). PMLR.
15. Lim, J., Kim, J., Kim, S., Park, N., & Yoon, S. (2024). MetaVers: Meta-Learned Versatile Representations for Personalized Federated Learning. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (pp. 7643-7652).
16. Niranjala, S. H., & Goyal, S. B. (2025). Improving Wine Quality Forecasts Using Dynamic Integral Neural Networks and Optimized Against Interference. International Journal on Smart & Sustainable Intelligent Computing, 2(1), 52–64.
17. Vadisetty, R., & Suyal, H. (2025). Smart Data Augmentation using Generative Adversarial Networks for Rare Oncological Disease Classification. International Journal on Smart & Sustainable Intelligent Computing, 2(2), 65-79.
18. Vinyals, O., Blundell, C., Lillicrap, T., Kavukcuoglu, K., & Wierstra, D. (2016). Matching networks for one shot learning. In Advances in neural information processing systems, 29.
