Transfer Faster, Price Smarter: Minimax Dynamic Pricing under Cross-Market Preference Shift

2025/05/22 by Zhang, Yi, Chen, Elynn, Yan, Yujun
#Applications (stat.AP) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Methodology (stat.ME)

paper · doi:10.48550/arxiv.2505.17203

Abstract

We study contextual dynamic pricing when a target market can leverage K auxiliary markets -- offline logs or concurrent streams -- whose mean utilities differ by a structured preference shift. We propose Cross-Market Transfer Dynamic Pricing (CM-TDP), the first algorithm that provably handles such model-shift transfer and delivers minimax-optimal regret for both linear and non-parametric utility models. For linear utilities of dimension d, where the difference between source- and target-task coefficients is s₀-sparse, CM-TDP attains regret O((d*K^-1+s₀)log T). For nonlinear demand residing in a reproducing kernel Hilbert space with effective dimension α, complexity β and task-similarity parameter H, the regret becomes O (K^{-2αβ/(2αβ+1)}T^1/(2αβ+1) + H^2/(2α+1)T^1/(2α+1)), matching information-theoretic lower bounds up to logarithmic factors. The RKHS bound is the first of its kind for transfer pricing and is of independent interest. Extensive simulations show up to 50% lower cumulative regret and 5 times faster learning relative to single-market pricing baselines. By bridging transfer learning, robust aggregation, and revenue optimization, CM-TDP moves toward pricing systems that transfer faster, price smarter.

Citations

• Prior-Aligned Meta-RL: Thompson Sampling with Learned Priors and Guarantees in Finite-Horizon MDPs
• Transfer Learning in Latent Contextual Bandits with Covariate Shift Through Causal Transportability
• Transition Transfer Q-Learning for Composite Markov Decision Processes
• Transfer Learning for Nonparametric Contextual Dynamic Pricing
• Deep Transfer Q-Learning for Offline Non-Stationary Reinforcement Learning
• Improved Algorithms for Contextual Dynamic Pricing
• Distributed Tensor Principal Component Analysis with Data Heterogeneity
• Dynamic Contextual Pricing with Doubly Non-Parametric Random Utility Models
• Data-Driven Knowledge Transfer in Batch Q^* Learning
• Transfer Learning for Contextual Multi-armed Bandits
• Transfer Q-learning
• Fully Online Meta-Learning Without Task Boundaries
• Towards Agnostic Feature-based Dynamic Pricing: Linear Policies vs Linear Valuation with Unknown Noise
• Multitask Learning and Bandits via Robust Statistics
• Policy Optimization Using Semi-parametric Models for Dynamic Pricing
• Dynamic Pricing in High-dimensions

Related

• Transfer Learning for Nonparametric Contextual Dynamic Pricing
• Smoothness-Adaptive Dynamic Pricing with Nonparametric Demand Learning
• Structured Dynamic Pricing: Optimal Regret in a Global Shrinkage Model
• Dynamic Contextual Pricing with Doubly Non-Parametric Random Utility Models
• Meta Dynamic Pricing: Transfer Learning Across Experiments