Newsvendor Model with Deep Reinforcement Learning
This is an incremental application of existing deep RL methods to a specific operations research problem.
The authors tackled the Newsvendor model, a mathematical problem for optimizing profit under probabilistic demand, using deep reinforcement learning to handle changing demand distributions across days of the week, and the agent learned optimal behavior consistent with analytical solutions.
I present a deep reinforcement learning (RL) solution to the mathematical problem known as the Newsvendor model, which seeks to optimize profit given a probabilistic demand distribution. To reflect a more realistic and complex situation, the demand distribution can change for different days of the week, thus changing the optimum behavior. I used a Twin-Delayed Deep Deterministic Policy Gradient agent (written as completely original code) with both an actor and critic network to solve this problem. The agent was able to learn optimal behavior consistent with the analytical solution of the problem, and could identify separate probability distributions for different days of the week and behave accordingly.