[Paper Review] Systematic Review on Reinforcement Learning in the Field of Fintech

Nadeem Malibari, Iyad Katib+1 more|arXiv (Cornell University)|Apr 29, 2023

FinTech, Crowdfunding, Digital FinanceBusiness, Management and Accounting3 citations

TL;DR

This systematic review evaluates reinforcement learning (RL) applications in fintech from 2018 onward using the PRISMA framework, analyzing RL's impact on portfolio optimization, risk management, and trading strategies. It finds RL-based approaches significantly outperform state-of-the-art algorithms in profitability, scalability, and decision-making accuracy across key fintech domains such as algorithmic trading, robo-advisory, and market making.

ABSTRACT

Applications of Reinforcement Learning in the Finance Technology (Fintech) have acquired a lot of admiration lately. Undoubtedly Reinforcement Learning, through its vast competence and proficiency, has aided remarkable results in the field of Fintech. The objective of this systematic survey is to perform an exploratory study on a correlation between reinforcement learning and Fintech to highlight the prediction accuracy, complexity, scalability, risks, profitability and performance. Major uses of reinforcement learning in finance or Fintech include portfolio optimization, credit risk reduction, investment capital management, profit maximization, effective recommendation systems, and better price setting strategies. Several studies have addressed the actual contribution of reinforcement learning to the performance of financial institutions. The latest studies included in this survey are publications from 2018 onward. The survey is conducted using PRISMA technique which focuses on the reporting of reviews and is based on a checklist and four-phase flow diagram. The conducted survey indicates that the performance of RL-based strategies in Fintech fields proves to perform considerably better than other state-of-the-art algorithms. The present work discusses the use of reinforcement learning algorithms in diverse decision-making challenges in Fintech and concludes that the organizations dealing with finance can benefit greatly from Robo-advising, smart order channelling, market making, hedging and options pricing, portfolio optimization, and optimal execution.

Motivation & Objective

• To examine the integration of reinforcement learning (RL) in financial technology (fintech) applications.
• To evaluate the performance of RL-based strategies in key fintech domains such as portfolio management and credit risk.
• To identify the strengths, limitations, and scalability of RL in real-world financial decision-making.
• To compare RL outcomes with state-of-the-art algorithms in terms of profitability, accuracy, and risk mitigation.
• To provide a comprehensive, evidence-based synthesis of recent advances in RL for fintech using a standardized review methodology.

Proposed method

• The study employs the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) framework to ensure methodological rigor and transparency.
• A systematic literature search was conducted focusing on peer-reviewed publications from 2018 onward in computational finance, AI, and machine learning venues.
• Inclusion and exclusion criteria were applied to filter studies based on relevance to RL in fintech, methodological soundness, and empirical evaluation.
• The review analyzes 31 pages of content, 15 figures, and 7 tables to extract insights on RL performance across multiple fintech applications.
• Data synthesis focused on comparative performance metrics, including prediction accuracy, profitability, and risk exposure.
• The review uses a four-phase flow diagram and checklist to ensure reproducibility and minimize bias in study selection and analysis.

Experimental results

Research questions

• RQ1 How does reinforcement learning improve decision-making in portfolio optimization compared to traditional methods?
• RQ2 What are the key performance metrics (e.g., profitability, risk reduction) of RL-based systems in fintech applications?
• RQ3 How do RL-based strategies compare to state-of-the-art algorithms in terms of scalability and robustness in financial environments?
• RQ4 What are the most prominent fintech applications of RL, and what challenges do they face in real-world deployment?
• RQ5 What are the limitations and risks associated with deploying RL in financial institutions?

Key findings

• RL-based strategies demonstrated superior performance in portfolio optimization, achieving higher risk-adjusted returns compared to conventional methods.
• In algorithmic trading and market making, RL models showed improved execution quality and reduced market impact.
• Robo-advisory systems using RL outperformed rule-based and supervised learning models in personalized investment recommendations.
• The use of RL in options pricing and hedging led to more accurate dynamic hedging strategies under market volatility.
• Despite strong performance, challenges related to interpretability, training stability, and regulatory compliance were consistently reported across studies.
• The review confirms that RL is scalable and effective in high-dimensional, sequential decision-making tasks common in fintech.