Abstract:Artificial intelligence is fundamentally transforming financial investment decision-making paradigms, with deep reinforcement learning (DRL) demonstrating significant application potential in domains such as robo-advisory services. Given that traditional portfolio optimization methods face significant challenges in effectively managing dynamic asset weight adjustments, this paper approaches the problem from the perspective of practical trading processes and develops a dynamic optimization model using deep reinforcement learning to achieve more effective asset allocation. The study's innovations are twofold: First, we propose a Sharpe ratio reward function specifically designed for Actor-Critic deep reinforcement learning algorithms, which optimizes portfolio performance by maximizing the average Sharpe ratio through random sampling and reinforcement learning algorithms during the training process; Second, we design deep neural networks that are specifically structured to meet asset optimization objectives. The study empirically evaluates the model using randomly selected constituent stocks from the CSI300 index and conducts comparative analyses against traditional approaches, including mean-variance optimization and risk parity strategies. Backtesting results demonstrate the dynamic optimization model's effectiveness in portfolio asset allocation, yielding enhanced risk reduction, superior risk-return metrics, and optimal performance across comprehensive evaluation criteria.

Portfolio Management using Deep Reinforcement Learning

A Novel Anti-Risk Method for Portfolio Trading Using Deep Reinforcement Learning

Dynamic Optimization of Portfolio Allocation Using Deep Reinforcement Learning

Deep Deterministic Portfolio Optimization

Deep reinforcement learning on a multi-asset environment for trading

High-Dimensional Stock Portfolio Trading with Deep Reinforcement Learning

Deep Robust Reinforcement Learning for Practical Algorithmic Trading

Adversarial Deep Reinforcement Learning in Portfolio Management

Application of Deep Reinforcement Learning on Automated Stock Trading

Deep Reinforcement Learning Approach for Trading Automation in The Stock Market

Advancing Investment Frontiers: Industry-grade Deep Reinforcement Learning for Portfolio Optimization

Application of Deep Q-Network in Portfolio Management

A Deep Reinforcement Learning Framework For Financial Portfolio Management

Multimodal Deep Reinforcement Learning for Portfolio Optimization

PortPortfolio Management Based on Deep Reinforcement Learning Method with Data Augment

Optimizing Portfolio Management and Risk Assessment in Digital Assets Using Deep Learning for Predictive Analysis

A Deep Reinforcement Learning Framework for the Financial Portfolio Management Problem

Evaluation of Deep Reinforcement Learning Based Stock Trading.

Deep Reinforcement Learning Strategies in Finance: Insights into Asset Holding, Trading Behavior, and Purchase Diversity

Practical Deep Reinforcement Learning Approach for Stock Trading

Application of deep reinforcement learning for Indian stock trading automation