Open Access

Cloud-Integrated Deep Reinforcement Learning for Adaptive Portfolio Risk Management in Complex Financial Systems

PDF

Lars Eklund ¹ ,

⁴ Lund University, Sweden

Abstract

The increasingly complex and interconnected nature of global financial markets has necessitated advanced computational techniques to predict, manage, and mitigate portfolio risk. Traditional risk management approaches, including Value-at-Risk (VaR) models and classical stochastic methods, often fail to capture dynamic, nonlinear dependencies between financial instruments under conditions of uncertainty and systemic stress (Holló, Kremer, & Lo Duca, 2012; Harford, Klasa, & Maxwell, 2014). Recent developments in machine learning, particularly deep reinforcement learning (DRL), offer promising alternatives by enabling adaptive, self-learning frameworks capable of dynamically adjusting portfolio strategies in real-time. This study presents a comprehensive analysis of an intelligent cloud framework designed to leverage DRL for dynamic portfolio risk prediction, integrating both macroeconomic indicators and micro-level asset behaviors (Mirza et al., 2025). The framework accommodates evolving market conditions, liquidity constraints, and geopolitical risks, providing a scalable and responsive solution for institutional and retail investors alike. Through extensive literature review, theoretical elaboration, and critical discussion, this paper contextualizes the proposed methodology within contemporary financial theory, geoeconomics, and systemic risk modeling. The study also identifies practical limitations, including data dependency, computational intensity, and interpretability challenges inherent in DRL-based systems. Ultimately, the research underscores the potential of cloud-integrated AI techniques in augmenting financial decision-making and mitigating systemic vulnerabilities in increasingly volatile global markets.

Keywords

Deep Reinforcement Learning, Portfolio Risk, Cloud Framework, Systemic Risk

References

Hodula, M., Janku, J., Malovana, S., & Ngo, N.A. (2024), “Geopolitical Risks and Their Impact on Global Macro-Financial Stability: Literature and Measurements”, Working Paper, No. 8, Česká národní banka, Prague, August.

Savona, P., & Regola, P. (2009), Il ritorno dello Stato padrone, Savoria Mannelli: Rubettino.

Thirlwell, M.P. (2010), The Return of Geoeconomics, Globalisation and National Security, The Lowy Institute, September.

Holló, D., Kremer, D., & Lo Duca, M. (2012), “CISS - A composite indicator of systemic stress in the financial system”, Working Paper Series, No 1426, ECB, Frankfurt am Main, March.

Harford, J., Klasa, S., & Maxwell, W.F. (2014), “Refinancing risk and cash holdings”, Journal of Finance, Vol. 69, Issue 3, Wiley Online Library, Hoboken, New Jersey, pp. 975-1012.

Sergiani, F., & Triulzi, U. (2016), Geofinanza e strategie di policy: verso un intelligence economica di contrasto, In Stati Generali dell’intelligence economica, Atti del convegno ed e-book, Roma: Tor Vergata University Press.

Savona, P. (2011), On the Macroeconomic Effects of Derivatives: Ten Lectures, Roma: LUISS University Press.

Hoffmann, P., Kremer, M., & Zaharia, S. (2019), “Financial integration in Europe through the lens of composite indicators”. Working Paper Series, No. 2319, ECB, Frankfurt am Main, September.

Sparke, M. (2018), Geoeconomics, Globalisation and the Limits of Economic Strategy in Statecraft: A Response to Vihma, Geopolitics, 23(1): 30-37.

Pegorer, P. (2011), Geografia dei sistemi finanziari, Trieste: Edizioni Università di Trieste.

Savona, P., & Oldani, C. (2005), Derivatives, fiscal policy and financial stability, ICFAI Journal of Derivatives, 2(3): 1-6.

IMF (2025a), External Sector Report: Global Imbalances in a Shifting World, International Monetary Fund, Washington, D.C., July.

Han, H., Linton, O., Oka, T., & Whang, Y.-J

. (2016), “The cross-quantilogram: Measuring quantile dependence and testing directional predictability between time series”, Journal of Econometrics, Vol. 193, No 1, Elsevier, Amsterdam, July, pp. 251-270.

Sergiani, F., & Triulzi, U. (2015), La geofinanza e l’impatto con i territori, In Marconi, M., Sellari P. (a cura di), Verso un nuovo paradigma geopolitico, Roma: Aracne, 279-302.

Sing, S., Razi, A., Endut, N., & Ramlee, H. (2007), Financial Market Developments and their Implication for Monetary Policy, BNM-BIS Conference Proceedings, Basilea, 13 agosto.

IMF (2025b), “Chapter 2. How to address the systemic part of liquidity risk”, in the Global financial stability report: Enhancing Resilience amid Uncertainty, International Monetary Fund, Washington, D.C., April.

Targetti, F. (2009), Globalizzazione e crisi economica. In Amato G. (a cura di), Governare l’economia globale, Firenze: Passigli. www.ferdinandotargetti.it/vgcf.htm.

Jiang, L., Bondell, H.D., & Wang, H.J. (2014), “Interquantile shrinkage and variable selection in quantile regression”, Computational statistics & data analysis, Vol. 69, Elsevier, Amsterdam, January, pp. 208-219.

Scholvin, S., & Wigell, M. (2018), Geo-Economics As Concept And Practice In International Relations: Surveying The State Of The Art, FIIA Working Paper, April 2018/102.

Sergiani, F. (2012), Geofinance: A New Approach to Link Finance and Territory, tesi di laurea.

Savona, P. (2004), Derivatives, money and real growth, Review of Financial Risk Management, WP: 106-120.

Gleditsch, N.P., Wallensteen, P., Eriksson, M., Sollenberg, M., & Strand, H. (2002), “Armed Conflict 1946-2001: A New Dataset”, Journal of Peace Research, Vol. 39, No 5, Sage Publishing, Thousand Oaks, California, September.

Sergiani, F., Triulzi, U. (2015), La geofinanza e l’impatto con i territori, In Marconi, M., Sellari P. (a cura di), Verso un nuovo paradigma geopolitico, Roma: Aracne, 279-302.

International Research Journal of Advanced Engineering and Technology

Cloud-Integrated Deep Reinforcement Learning for Adaptive Portfolio Risk Management in Complex Financial Systems

Abstract

Keywords

References

Most read articles by the same author(s)

Similar Articles