Graph-Attentive MAPPO for Dynamic Retail Pricing

TL;DR

This paper systematically evaluates multi-agent reinforcement learning for retail pricing, demonstrating that graph-attention mechanisms improve performance and stability in multi-product dynamic pricing scenarios.

Contribution

It introduces a graph-attention-augmented MAPPO variant that leverages product interactions, enhancing multi-product pricing strategies over standard MAPPO.

Findings

MAPPO is a robust foundation for retail price control.

MAPPO+GAT improves performance by sharing information across products.

Graph-integrated MARL offers scalable, stable solutions for dynamic pricing.

Abstract

Dynamic pricing in retail requires policies that adapt to shifting demand while coordinating decisions across related products. We present a systematic empirical study of multi-agent reinforcement learning for retail price optimization, comparing a strong MAPPO baseline with a graph-attention-augmented variant (MAPPO+GAT) that leverages learned interactions among products. Using a simulated pricing environment derived from real transaction data, we evaluate profit, stability across random seeds, fairness across products, and training efficiency under a standardized evaluation protocol. The results indicate that MAPPO provides a robust and reproducible foundation for portfolio-level price control, and that MAPPO+GAT further enhances performance by sharing information over the product graph without inducing excessive price volatility. These results indicate that graph-integrated MARL provides a more scalable and stable solution than independent learners for dynamic retail pricing, offering practical advantages in multi-product decision-making.