Deep Reinforcement Learning for Trading: Using Gradient Ascent to Maximize Sharpe Ratio Peter Foy - 12 Apr 2022