A Comparative Study of Dynamic Programming and Reinforcement Learning in Finite Horizon Dynamic Pricing — ThinkLLM