Fast and Feasible: Permutation-based Constrained Reranking for Revenue Maximization

2026-06-26

AI Summary

This work addresses the challenge of revenue optimization in e-commerce search reranking, where maximizing platform revenue alone can degrade user experience or induce fraudulent behavior. To balance revenue with multidimensional constraints such as relevance, the problem is formulated as a constrained integer linear program (ILP). The authors propose PermR, a lightweight pairwise-swap heuristic algorithm that efficiently approximates optimal revenue while strictly satisfying constraints and meeting low-latency requirements. Offline experiments demonstrate that PermR achieves approximately 63% of the revenue gain attainable by the full ILP solution. Furthermore, a 14-day online A/B test involving 56 million queries shows a statistically significant 2% increase in revenue, confirming the method’s effectiveness and practicality in real-world deployment.

Abstract

Search and recommender systems have produced highly relevant search results. A natural next step in the development of such systems in e-commerce is to rerank these results to increase the platform's revenue from paid promotion products. However, maximizing revenue alone may degrade the user experience by reducing relevance or increasing fraud risk. To avoid this, we state the reranking problem as an integer linear program ($ILP$) that maximizes revenue subject to per-query constraints on other metrics, e.g., relevance. Since solving $ILP$ exactly for every query is slow for deployment to the online service, we propose a lightweight permutation-based reranking approximation algorithm PermR. At each step, the algorithm selects a pair of neighboring items and swaps them to either improve the objective or repair a violated constraint. We evaluate PermR across multiple categories of a large classified platform in offline and online settings. PermR achieves about 63\% of the ILP revenue improvement, within production latency limits, preserving all constraints. In a 14-day online A/B test over 56 million search queries, PermR increased revenue by $2$\%.

Problem

• reranking
• revenue maximization
• per-query constraints
• relevance
• e-commerce

Innovation

• permutation-based reranking
• revenue maximization
• constrained optimization
• integer linear programming
• online A/B testing