– The research proposes a new automated cryptocurrency trading system integrated with DRL.
– The experimental analysis of the model showed exceptional results, surpassing similar works.
– Development of a stable, accurate, and robust automated trading system.
– Implementation of the proximal policy optimization (PPO) algorithm in an automated trading system.
The paper discusses the implementation of a deep reinforcement learning algorithm in an automated cryptocurrency trading system. It aims to develop a stable and accurate system to predict price movements and maximize investment returns.
– Humans have limitations in terms of availability and rational thinking.
– Humans tend to be their own greatest enemy due to emotions.
– Data acquisition from Binance exchange REST API v3
– RL and DNN architecture for model training
– Resolves human hindrance in automated trading.
– Demonstrates exceptional results surpassing similar works.
– The research model outperformed the baseline buy and hold method.
– The research model exceeded models of other similar works.
– Proposed implementation of PPO algorithm in an automated trading system.
– Aims to develop a stable, accurate, and robust trading system using DRL.
– The paper proposes implementing the PPO algorithm in an automated trading system.
– The research aims to develop a stable and accurate trading system using DRL.
