Survey on Intelligent Trust-Weighted Reinforcement Learning for Financial Advisory Systems

Abstract

Financial advisory systems are increasingly adopting intelligent computational methods to assist investors in making informed financial decisions. Traditional advisory systems rely on predefined rules and static financial models that lack adaptability in dynamic market environments. Reinforcement Learning (RL), a branch of artificial intelligence focused on learning optimal strategies through interaction with an environment, has emerged as a promising approach for adaptive financial decision-making. This paper presents a survey of reinforcement learning based financial advisory systems that learn investment strategies through reward-driven optimization. These systems dynamically adjust portfolio allocations, trading actions, and investment recommendations based on real-time market feedback. Reinforcement learning models such as Q-learning, policy gradient methods, and deep reinforcement learning have been widely explored for portfolio optimization and trading strategy development. Additionally, trust-aware mechanisms and explainable frameworks are increasingly integrated into RL systems to improve transparency and reliability. This survey reviews existing reinforcement learning techniques applied in financial advisory systems, discusses the challenges of market volatility and system interpretability, and identifies research gaps for future development of intelligent and trustworthy RL-based financial advisory platforms.