Published Paper Details:

Stock trading techniques play the essential aspect of functioning as a key factor when it comes to investment. In this case Paper, It proposes a way of learning a stock trading strategy that maximizes return on investment using Deep Reintorcement.Researchers developed ways to buy and sell using deep reinforcement learning agents using ensemble trading strategies for actor-critical-based models on these 5 Algorithms: Proximal Policy Optimization (PPO), Actor-Critic Gain (A2C), Deep Deterministic Policy Gradient (DDPG), Twin-Delayed Deep Deterministic Policy Gradient (TD3), Soft Actor-Critic(SAC). Ensemble strategy inherits and integrates the excellent features of these five Algorithms, thereby robustly adapting to the exceptional market situation.To avoid immense memory usage in training networks with an area of continuous movement, For processing massive amounts of data, researchers employ the load-on-demand approach, 30 Dow Jones Stocks with enough liquidity are used to test the algorithms.The overall performance of the trading agency was evaluated with significant enhancement of the Reinforcement algorithm and compared to both the Dow Jones trading stock index and the traditional minimum deviation portfolio allocation procedure. The proposed deep ensemble strategy outperforms baselines and five algorithms in trade-weighted returns by the Sharpe ratio.