Deep Reinforcement Learning methods for Automation Forex Trading

In Forex market, designing effective strategies are a critical role in investment. However, it is a challenging task due to its inherent characteristics, which include high volatility, trend, noise, and market shocks. In this paper, we propose four actor-critic-based algorithms: Proximal Policy Optimization (PPO), Actor-Critic using Kronecker-Factored Trust Region (ACKTR), Deep Deterministic Policy Gradient (DDPG), and Twin Delayed DDPG (TD3). It employs deep reinforcement schemes to learn a stock trading strategy by maximizing investment return. Besides, the ensemble trading strategy was combined with four algorithms to find the best method. We use our algorithms to test the 30 forex currencies.