reinforcement learning Beats Markov Decision Process stock trading mdp bits Big Reference Policy Deterministic Policies Stochastic Policies Portfolio Optimization Forward and Inverse Optimization One-step Reward Market Impact Penalty Fee Penalty Risk Penalty Instantaneous Rewards Initial Conditions Terminal Conditions The Next-period Portfolio Asset Returns Model One-period Investment Self-financing Condition Portfolio Model Optimal Portfolio Quantitative Trading Bellman Equation Value Function Stochastic Policy Deterministic Policy Markov Decision Processes Multiobjective Optimization Multiobjective Evolutionary Algorithms Evolutionary Algorithms modern portfolio theory low volatility funds standardization stock selection threejs Sharpe Ratio Genetic Algorithm Exploitation exploration Graphics Portfolio finance Economics 3D