Computational Performance of Deep Reinforcement Learning to find Nash Equilibria

Abstract

We test the performance of deep deterministic policy gradient (DDPG), a deep reinforcement learning algorithm, able to handle continuous state and action spaces, to learn Nash equilibria in a setting where firms compete in prices. These algorithms are typically considered model-free because they do not require transition probability functions (as in e.g., Markov games) or predefined functional forms. Despite being model-free, a large set of parameters are utilized in various steps of the algorithm. These are e.g., learning rates, memory buffers, state-space dimensioning, normalizations, or noise decay rates and the purpose of this work is to systematically test the effect of these parameter configurations on convergence to the analytically derived Bertrand equilibrium. We find parameter choices that can reach convergence rates of up to 99%. The reliable convergence may make the method a useful tool to study strategic behavior of firms even in more complex settings. Keywords: Bertrand Equilibrium, Competition in Uniform Price Auctions, Deep Deterministic Policy Gradient Algorithm, Parameter Sensitivity Analysis

arXiv comment48 pages + 9 figures, comments welcome!

References in corpus (9)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Continuous control with deep reinforcement learning
• Addressing Function Approximation Error in Actor-Critic Methods
• Mastering Atari, Go, Chess and Shogi by Planning with a Learned Model
• Learning to Communicate with Deep Multi-Agent Reinforcement Learning
• Learning values across many orders of magnitude
• No-regret Dynamics and Fictitious Play
• (Machine) Learning from the COVID-19 Lockdown about Electricity Market Performance with a Large Share of Renewables
• A Reinforcement Learning Approach for the Continuous Electricity Market of Germany: Trading from the Perspective of a Wind Park Operator