reinforcement learning, Markov decision process, dynamic
programming, machine learning, Monte Carlo, Q-Learning,
REINFORCE, Actor-critic, expected utility, mean-variance,
conditional value at risk, CVaR, risk of ruin,
chance-constraint