Autonomous Financial Decision Systems Using Reinforcement Learning for Dynamic Credit Risk Management

Dr Somasundaram Krishnan; Dr Somasundaram Krishnan

doi:10.15662/IJSRAT.2026.0903010

Autonomous Financial Decision Systems Using Reinforcement Learning for Dynamic Credit Risk Management

Dr Somasundaram Krishnan

Professor, Department of Computer Science and Engineering, Sri Muthukumaran Institute of Technology, Chennai, India

Abstract

The systems of autonomous financial decisions are becoming more and more demanded to control credit risk within dynamic and uncertain credit lending conditions. In the current paper, a new Autonomous Financial Decision Framework (AFDF) is proposed, applying the principles of the Reinforcement Learning (RL), real-time financial data analysis, predictive risk modeling and adaptive decision-making to address dynamical credit risk. The framework has six important parts: data acquisition and preprocessing, risk profiling of the borrowers, state representation, the optimization of the policy, by means of reinforcement learning, independent credit decision-making and continuous monitoring with feedback. The proposed system is a dynamic system that represents the borrower and market-state(s) as dynamic systems and engages an RL agent in selecting the promise credit behavior, which can be accepting credit, rejecting credit, raising or lowering credit limits, risk-based intervention. A reward scheme balances between profitability, credit quality, default risk and regulatory limitation and enables the system to continually optimize its issuances as the operating financial performance changes. The model is designed to overcome lack of varying other inactive credit-scoring models by adapting itself to the variable borrower behavior and market forces. It has the potential to aid financial institutions to enhance the consistency of decisions, possible credit losses as well as operational efficiency. The specified solution provides the scalable foundation of autonomous, adaptive and risk-aware financial decisions.

Article Information

Journal	International Journal of Science, Research and Technology
Volume (Issue)	Vol. 9 No. 3 (2026): International Journal of Science, Research and Technology (IJSRAT)
DOI	https://doi.org/10.15662/IJSRAT.2026.0903010
Pages	895-905
Published	May 14, 2026
Copyright	All rights reserved
Open Access	This work is licensed under a Creative Commons Attribution 4.0 International License.
How to Cite	Dr Somasundaram Krishnan (%2026). Autonomous Financial Decision Systems Using Reinforcement Learning for Dynamic Credit Risk Management. International Journal of Science, Research and Technology , Vol. 9 No. 3 (2026): International Journal of Science, Research and Technology (IJSRAT) , pp. 895-905. https://doi.org/10.15662/IJSRAT.2026.0903010

References

[1] H. Ayari and R. Guetari, “Integrating genetic algorithms and ensemble learning for improved and transparent credit scoring,” in Proc. Int. Conf. Business Information Systems, 2025, pp. 225–238, doi: 10.1007/978-3-031-94193-1_17.
[2] C. S. Kubam, J. Duggirala, S. Vishnubhai Sheta, S. K. Mogali, U. Lakhina, and H. Kaur, “AI-Driven Credit Risk Assessment in Digital Finance Using Feature Optimization Deep Q Learning,” in Proc. 2025 International Conference on Computational Intelligence and Knowledge Economy (ICCIKE), 2025, pp. 210–216, IEEE.
[3] R. Bhandary and B. K. Ghosh, “Credit card default prediction: An empirical analysis on predictive performance using statistical and machine learning methods,” Journal of Risk and Financial Management, vol. 18, no. 1, Art. no. 23, 2025, doi: 10.3390/jrfm18010023.
[4] M. Bai, Y. Zheng, and Y. Shen, “Gradient boosting survival tree with applications in credit scoring,” Journal of the Operational Research Society, vol. 73, no. 1, pp. 39–55, 2022, doi: 10.1080/01605682.2021.1919035.
[5] M. J. Ariza-Garzón, J. Arroycite o, A. Caparrini, and M.-J. Segovia-Vargas, “Explainability of a machine learning granting scoring model in peer-to-peer lending,” IEEE Access, vol. 8, pp. 64873–64890, 2020, doi: 10.1109/ACCESS.2020.2984412.
[6] W. Bao, L. Ning, and K. Yue, “Integration of unsupervised and supervised machine learning algorithms for credit risk assessment,” Expert Systems with Applications, vol. 128, pp. 301–315, 2019, doi: 10.1016/j.eswa.2019.02.033.
[7] S. Bhatore, L. Mohan, and Y. R. Reddy, “Machine learning techniques for credit risk evaluation: A systematic literature review,” Journal of Banking and Financial Technology, vol. 4, no. 1, pp. 111–138, 2020, doi: 10.1007/s42786-020-00020-3.
[8] A. F. Atiya, “Bankruptcy prediction for credit risk using neural networks: A survey and new results,” IEEE Transactions on Neural Networks, vol. 12, no. 4, pp. 929–935, 2001, doi: 10.1109/72.935101.
[9] R. Bhandary, S. S. Shenoy, A. Shetty, and A. D. Shetty, “Education loan repayment: A systematic literature review,” Journal of Financial Services Marketing, vol. 29, no. 4, pp. 1365–1376, 2024, doi: 10.1057/s41264-023-00248-2.
[10] M. K. Lessmann, B. Baesens, H.-V. Seow, and L. C. Thomas, “Benchmarking state-of-the-art classification algorithms for credit scoring: An update of research,” European Journal of Operational Research, vol. 247, no. 1, pp. 124–136, 2015.
[11] N. Khandani, A. Kim, and A. Lo, “Consumer credit-risk models via machine-learning algorithms,” Journal of Banking & Finance, vol. 34, no. 11, pp. 2767–2787, 2010.
[12] S. Chen, Y. Chen, and X. Li, “Credit score prediction using genetic algorithm-LSTM technique,” Journal of Intelligent & Fuzzy Systems, 2022.
[13] W. Zhang, X. He, and J. Zhang, “A new deep learning ensemble credit risk evaluation model with an improved synthetic minority oversampling technique,” International Journal of Information Management Data Insights, 2021.