MultiSlot ReRanker: A Generic Model-based Re-Ranking Framework in Recommendation Systems

TL;DR

This paper introduces MultiSlot ReRanker, a scalable, model-based re-ranking framework that optimizes relevance, diversity, and freshness in recommendation systems, achieving significant offline performance improvements.

Contribution

It presents a novel, efficient re-ranking algorithm with theoretical analysis and a versatile simulation environment for benchmarking various algorithms.

Findings

Achieved +6% to +10% offline AUC improvements.

Developed a linear-time greedy re-ranking algorithm.

Created a flexible simulation platform for algorithm testing.

Abstract

In this paper, we propose a generic model-based re-ranking framework, MultiSlot ReRanker, which simultaneously optimizes relevance, diversity, and freshness. Specifically, our Sequential Greedy Algorithm (SGA) is efficient enough (linear time complexity) for large-scale production recommendation engines. It achieved a lift of $+6\%$ to $ +10\%$ offline Area Under the receiver operating characteristic Curve (AUC) which is mainly due to explicitly modeling mutual influences among items of a list, and leveraging the second pass ranking scores of multiple objectives. In addition, we have generalized the offline replay theory to multi-slot re-ranking scenarios, with trade-offs among multiple objectives. The offline replay results can be further improved by Pareto Optimality. Moreover, we've built a multi-slot re-ranking simulator based on OpenAI Gym integrated with the Ray framework. It can be easily configured for different assumptions to quickly benchmark both reinforcement learning and supervised learning algorithms.