# Pairwise Learning to Rank by Neural Networks Revisited: Reconstruction, Theoretical Analysis and Practical Performance

arXiv: https://arxiv.org/abs/1909.02768 · HTML: https://mlsanity.com/paper/1909.02768/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 Sep 2019.

## Built on by (most cited first, 13 of 13)

- [GNNRank: Learning Global Rankings from Pairwise Comparisons via Directed Graph Neural Networks](https://mlsanity.com/paper/2202.00211/) (2022)
- [Offline Model-Based Optimization by Learning to Rank](https://mlsanity.com/paper/2410.11502/) (2024)
- [Learning to Compare Longitudinal Images](https://mlsanity.com/paper/2304.02531/) (2023)
- [Multi-task Bioassay Pre-training for Protein-ligand Binding Affinity Prediction](https://mlsanity.com/paper/2306.04886/) (2023)
- [Rxn Hypergraph: a Hypergraph Attention Model for Chemical Reaction Representation](https://mlsanity.com/paper/2201.01196/) (2022)
- [Fair Interpretable Representation Learning with Correction Vectors](https://mlsanity.com/paper/2202.03078/) (2022)
- [The Evolution of Reranking Models in Information Retrieval: From Heuristic Methods to Large Language Models](https://mlsanity.com/paper/2512.16236/) (2025)
- [Fine-grained Analysis of Non-parametric Estimation for Pairwise Learning](https://mlsanity.com/paper/2305.19640/) (2023)
- [Mobile Supply: The Last Piece of Jigsaw of Recommender System](https://mlsanity.com/paper/2308.03855/) (2023)
- [Mitigating Judgment Preference Bias in Large Language Models through Group-Based Polling](https://mlsanity.com/paper/2510.08145/) (2025)
- [Evaluating Local Model-Agnostic Explanations of Learning to Rank Models with Decision Paths](https://mlsanity.com/paper/2203.02295/) (2022)
- [Perturbation-based Active Learning for Question Answering](https://mlsanity.com/paper/2311.02345/) (2023)
- [Preference-based learning for news headline recommendation](https://mlsanity.com/paper/2506.06334/) (2025)

## Abstract (arXiv)

We present a pairwise learning to rank approach based on a neural net, called DirectRanker, that generalizes the RankNet architecture. We show mathematically that our model is reflexive, antisymmetric, and transitive allowing for simplified training and improved performance. Experimental results on the LETOR MSLR-WEB10K, MQ2007 and MQ2008 datasets show that our model outperforms numerous state-of-the-art methods, while being inherently simpler in structure and using a pairwise approach only.