arXiv q-fin· Hrishikesh Singh, Aarti Sharma, Millie Pant·· 1 days agoAI Score33
Deep Reinforcement Learning Methods for Stock Trading: Actor-Critic Agents with Rich Financial Context
Financial Context-Rich Strategy for Actor-Critic based Agents: A Deep Reinforcement Learning Approach for Stock Trading
AI summary
This study proposes a deep reinforcement learning stock trading method that integrates a sentiment analysis component with deep injection. It designs two types of Actor-Critic agent strategies: A2C and PPO for policy optimization only, and DDPG, TD3, and SAC that incorporate Q-learning principles. The study finds that agents combining Q-learning, when deep-injected with sentiment information, achieve significant and stable performance improvements in profit maximization and risk-averse trading.
Source:arXiv q-fin · arxiv.org
Academic papers#Papers / Research#Machine Learning#Large Language Model Applications#Risk Management#Backtesting#US Stocks