arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2406.02437econ.GNq-fin.EC
收藏

动态定价中基于深度强化学习的算法合谋

Algorithmic Collusion in Dynamic Pricing with Deep Reinforcement Learning

Shidi Deng, Maximilian Schiffer, Martin Bichler

AI总结:

本研究探讨了动态定价中强化学习算法引发算法合谋的风险,通过对比TQL与多种DRL算法在Bertrand寡头市场中的表现,发现合谋程度取决于算法类型与市场环境特征,且PPO对合谋结果较不敏感。

摘要:

如今,B2C(企业对消费者)领域的很大一部分基于 Amazon 和 Alibaba 等在线平台,并使用人工智能(AI)制定定价策略。这引发了争论:定价算法是否会在未被明确设计如此的情况下,默契地合谋设定超竞争价格。本研究通过考察在竞争市场中使用强化学习(RL)算法决定定价策略时的合谋风险来解决这些问题。该领域先前的研究集中于 Tabular Q-learning(TQL),并对基于学习的算法是否会导致超竞争价格产生了对立观点。我们的工作通过提供更细致的数值研究来推进这一持续讨论,该研究超越了 TQL,额外涵盖了 off-policy 和 on-policy 深度强化学习(DRL)算法。我们研究了多种 Bertrand 寡头垄断变体,并表明算法合谋取决于所使用的算法。在我们的实验中,与 DRL 算法相比,TQL 表现出更高的合谋和价格分散现象。我们表明,合谋的严重程度不仅取决于所使用的算法,还取决于市场环境的特征。我们进一步发现,与其他最先进的 DRL 算法相比,Proximal Policy Optimization 似乎对合谋结果不太敏感。

英文摘要:

Nowadays, a significant share of the Business-to-Consumer sector is based on online platforms like Amazon and Alibaba and uses Artificial Intelligence for pricing strategies. This has sparked debate on whether pricing algorithms may tacitly collude to set supra-competitive prices without being explicitly designed to do so. Our study addresses these concerns by examining the risk of collusion when Reinforcement Learning algorithms are used to decide on pricing strategies in competitive markets. Prior research in this field focused on Tabular Q-learning (TQL) and led to opposing views on whether learning-based algorithms can lead to supra-competitive prices. Our work contributes to this ongoing discussion by providing a more nuanced numerical study that goes beyond TQL by additionally capturing off- and on-policy Deep Reinforcement Learning (DRL) algorithms. We study multiple Bertrand oligopoly variants and show that algorithmic collusion depends on the algorithm used. In our experiments, TQL exhibits higher collusion and price dispersion phenomena compared to DRL algorithms. We show that the severity of collusion depends not only on the algorithm used but also on the characteristics of the market environment. We further find that Proximal Policy Optimization appears to be less sensitive to collusive outcomes compared to other state-of-the-art DRL algorithms.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑