Harm Mitigation in Recommender Systems under User Preference Dynamics

Jerry Chee, Shankar Kalyanaraman, Sindhu Kiranmai Ernala, Udi Weinsberg, Sarah Dean +1 more
6/14/2024
cs.IRcs.CYcs.LG

No official code found. Best implementation: jerry-chee/HarmMitigation (1 star, last commit 2024-06-18).

Abstract

We consider a recommender system that takes into account the interplay between recommendations, the evolution of user interests, and harmful content. We model the impact of recommendations on user behavior, particularly the tendency to consume harmful content. We seek recommendation policies that establish a tradeoff between maximizing click-through rate (CTR) and mitigating harm. We establish conditions under which the user profile dynamics have a stationary point, and propose algorithms for finding an optimal recommendation policy at stationarity. We experiment on a semi-synthetic movie recommendation setting initialized with real data and observe that our policies outperform baselines at simultaneously maximizing CTR and mitigating harm.

View on arXivView PDF

Code Implementations(1)

jerry-chee/HarmMitigationCommunity100%
10PythonJun 9, 20242 years ago

Similar Papers

Modeling Churn in Recommender Systems with Aggregated Preferences

85% match2025

System-2 Recommenders: Disentangling Utility and Engagement in Recommendation Systems via Temporal Point-Processes

83% match2024

Ensuring User-side Fairness in Dynamic Recommender Systems

82% match2023

The Nah Bandit: Modeling User Non-compliance in Recommendation Systems

82% match1 repo2024

Modeling Recommender Ecosystems: Research Challenges at the Intersection of Mechanism Design, Reinforcement Learning and Generative Models

82% match2023

User-Creator Feature Polarization in Recommender Systems with Dual Influence

81% match2024

Cite this paper

@article{chee2024harm,
  title  = {Harm Mitigation in Recommender Systems under User Preference Dynamics},
  author = {Jerry Chee and Shankar Kalyanaraman and Sindhu Kiranmai Ernala and Udi Weinsberg and Sarah Dean and Stratis Ioannidis},
  year   = {2024},
  eprint = {2406.09882},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2406.09882v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2406.09882.svg)](https://codeofpaper.com/paper/2406.09882)

Discussion