# Learning to Rank in the Position Based Model with Bandit Feedback

arXiv: https://arxiv.org/abs/2004.13106 · HTML: https://mlsanity.com/paper/2004.13106/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 27 Apr 2020.

## Built on by (most cited first, 10 of 10)

- [Optimal Design for Human Preference Elicitation](https://mlsanity.com/paper/2404.13895/) (2024)
- [A Map of Bandits for E-commerce](https://mlsanity.com/paper/2107.00680/) (2021)
- [Adversarial Attacks on Online Learning to Rank with Click Feedback](https://mlsanity.com/paper/2305.17071/) (2023)
- [Contextual bandits with concave rewards, and an application to fair ranking](https://mlsanity.com/paper/2210.09957/) (2022)
- [Bandit Learning to Rank with Position-Based Click Models: Personalized and Equal Treatments](https://mlsanity.com/paper/2311.04528/) (2023)
- [Overcoming Prior Misspecification in Online Learning to Rank](https://mlsanity.com/paper/2301.10651/) (2023)
- [UniRank: Unimodal Bandit Algorithm for Online Ranking](https://mlsanity.com/paper/2208.01515/) (2022)
- [SmartChoices: Augmenting Software with Learned Implementations](https://mlsanity.com/paper/2304.13033/) (2023)
- [MetaSynth: Multi-Agent Metadata Generation from Implicit Feedback in Black-Box Systems](https://mlsanity.com/paper/2510.01523/) (2025)
- [Design and Evaluation of Whole-Page Experience Optimization for E-commerce Search](https://mlsanity.com/paper/2602.02514/) (2026)

## Abstract (arXiv)

Personalization is a crucial aspect of many online experiences. In particular, content ranking is often a key component in delivering sophisticated personalization results. Commonly, supervised learning-to-rank methods are applied, which suffer from bias introduced during data collection by production systems in charge of producing the ranking. To compensate for this problem, we leverage contextual multi-armed bandits. We propose novel extensions of two well-known algorithms viz. LinUCB and Linear Thompson Sampling to the ranking use-case. To account for the biases in a production environment, we employ the position-based click model. Finally, we show the validity of the proposed algorithms by conducting extensive offline experiments on synthetic datasets as well as customer facing online A/B experiments.