English

Smoothness-Adaptive Dynamic Pricing with Nonparametric Demand Learning

Machine Learning 2023-11-02 v2 Machine Learning Econometrics

Abstract

We study the dynamic pricing problem where the demand function is nonparametric and H\"older smooth, and we focus on adaptivity to the unknown H\"older smoothness parameter β\beta of the demand function. Traditionally the optimal dynamic pricing algorithm heavily relies on the knowledge of β\beta to achieve a minimax optimal regret of O~(Tβ+12β+1)\widetilde{O}(T^{\frac{\beta+1}{2\beta+1}}). However, we highlight the challenge of adaptivity in this dynamic pricing problem by proving that no pricing policy can adaptively achieve this minimax optimal regret without knowledge of β\beta. Motivated by the impossibility result, we propose a self-similarity condition to enable adaptivity. Importantly, we show that the self-similarity condition does not compromise the problem's inherent complexity since it preserves the regret lower bound Ω(Tβ+12β+1)\Omega(T^{\frac{\beta+1}{2\beta+1}}). Furthermore, we develop a smoothness-adaptive dynamic pricing algorithm and theoretically prove that the algorithm achieves this minimax optimal regret bound without the prior knowledge β\beta.

Keywords

Cite

@article{arxiv.2310.07558,
  title  = {Smoothness-Adaptive Dynamic Pricing with Nonparametric Demand Learning},
  author = {Zeqi Ye and Hansheng Jiang},
  journal= {arXiv preprint arXiv:2310.07558},
  year   = {2023}
}

Comments

Minor typo errors corrected in the latest version

R2 v1 2026-06-28T12:47:28.638Z