Dynamic pricing under competition using reinforcement learning

Abstract Dynamic pricing is considered a possibility to gain an advantage over competitors in modern online markets. The past advancements Reinforcement Learning (RL) provided more capable algorithms that can be used solve problems. In this paper, we study the performance of Deep Q-Networks (DQN) and Soft Actor Critic (SAC) different market models. We consider tractable duopoly settings, where optimal solutions derived by dynamic programming techniques for verification, as well oligopoly which are usually intractable due curse dimensionality. find both provide reasonable results, while SAC performs better than DQN. Moreover, show under certain conditions, RL forced into collusion their without direct communication.