pythonpaperrlddpgafter-work
Introducing Practical Deep Reinforcement Learning Approach for Stock Trading for after-work quants
Liu, Xiong, Zhong, Yang and Walid (2018) train DDPG on DJIA 30 daily prices (2009–2018). State is prices, holdings, cash. Beats min-variance and DJIA on 2016–2018 Sharpe in-sample-style test. Plain-language intro. Not investment advice.