Reinforcement learning for trading

Reinforcement learning applied to quantitative trading has three distinct components. First of all, reinforcement learning agents require an environment in which to learn. A cryptocurrency perpetual future and spot environment are presented here. Secondly the specific reinforcement learning algorithm that is used as basis for the agent, value-based and policy-based learning algorithms are presented here. Thirdly, the marrying of the agent and the environment is presented in learning scripts where the agent will learn and present validation results. These learning scripts are often specific to agent and environment, so they are included in the agents folder.