Cryptocurrency trading with DRL

A realistic Gymnasium environment for Crypto Trading using Deep Reinforcement Learning (DRL) with stochastic slippage and market impact models.

Deep Reinforcement Learning project implementing and evaluating trading agents using historical market data. The goal is to optimize trading strategies with DRL algorithms, balancing profit maximization and risk management in crypto markets.

Developed a DRL-based crypto trading model using Proximal Policy Optimization (PPO) in the FinRL framework. Trained the agent on historical data to optimize portfolio returns, incorporating feature engineering and financial indicators for improved performance.