Construction and Performance Algorithm Optimization of Financial Dynamic Pricing Model Based on Reinforcement Learning
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
To solve model construction and algorithm performance optimization under multi-constraint dynamic environments in financial dynamic pricing, this study proposes a scheme integrating an improved artificial bee colony algorithm with reinforcement learning (RL). In the RL agent’s exploration-exploitation, the Pareto dominance criterion is introduced to optimize state space search, enhancing financial market dynamics exploration and avoiding premature convergence from incomplete market information. In the model parameter iterative optimization phase (employed bee phase), key features (e.g., market transaction status, demand elasticity) are identified in real time to adaptively select parameter search equations and constraint strategies (e.g., pricing range, risk-return constraints), guiding rapid convergence to feasible pricing ranges. In the performance improvement phase (onlooker bee phase), RL-accumulated global optimal pricing strategies and return feedback guide further parameter exploration, boosting the algorithm’s development capability and performance stability. Experiments on simulated financial datasets and 20 CEC 2006 test functions show the scheme balances constraint satisfaction and optimization efficiency. Applied to actual financial products (e.g., wealth management, insurance), numerical results confirm its effectiveness in improving pricing accuracy, return stability, and market adaptability, supporting financial dynamic pricing.
Fields of Study & Subjects
Metaheuristic Optimization Algorithms Research
Stock Market Forecasting Methods
Advanced Multi-Objective Optimization Algorithms
Publication details
- Journal / source
- IEEE International Conference on Power, Electronics and Green Energy (ICPEGE)
- Pages
- 469–476
- Publication date
- 28 January 2026
- Language
- EN