UMRE: A Unified Monotonic Transformation for Ranking Ensemble in Recommender Systems

TL;DR

This paper introduces UMRE, a novel ensemble ranking framework for recommender systems that uses neural networks to learn monotonic transformations, eliminating manual tuning and improving personalization and Pareto efficiency.

Contribution

UMRE replaces handcrafted transformations with neural networks, introduces a Pareto optimality strategy, and demonstrates superior performance on recommendation datasets and online tests.

Findings

Achieves better ranking performance than traditional methods

Eliminates manual tuning of transformation functions and weights

Demonstrates strong generalization in online A/B tests

Abstract

Industrial recommender systems commonly rely on ensemble sorting (ES) to combine predictions from multiple behavioral objectives. Traditionally, this process depends on manually designed nonlinear transformations (e.g., polynomial or exponential functions) and hand-tuned fusion weights to balance competing goals -- an approach that is labor-intensive and frequently suboptimal in achieving Pareto efficiency. In this paper, we propose a novel Unified Monotonic Ranking Ensemble (UMRE) framework to address the limitations of traditional methods in ensemble sorting. UMRE replaces handcrafted transformations with Unconstrained Monotonic Neural Networks (UMNN), which learn expressive, strictly monotonic functions through the integration of positive neural integrals. Subsequently, a lightweight ranking model is employed to fuse the prediction scores, assigning personalized weights to each prediction objective. To balance competing goals, we further introduce a Pareto optimality strategy that adaptively coordinates task weights during training. UMRE eliminates manual tuning, maintains ranking consistency, and achieves fine-grained personalization. Experimental results on two public recommendation datasets (Kuairand and Tenrec) and online A/B tests demonstrate impressive performance and generalization capabilities.