Rank List Sensitivity of Recommender Systems to Interaction Perturbations

TL;DR

This paper introduces Rank List Sensitivity (RLS), a measure of how small data perturbations affect recommender system outputs, revealing high instability especially for low-accuracy users and proposing CASPER to identify minimal perturbations.

Contribution

The paper proposes RLS as a new stability measure for recommender systems and introduces CASPER to systematically find perturbations that increase model instability.

Findings

Recommender models are highly sensitive to minor perturbations.

CASPER can identify minimal perturbations that drastically change recommendations.

Low-accuracy users experience more unstable recommendations under perturbations.

Abstract

Prediction models can exhibit sensitivity with respect to training data: small changes in the training data can produce models that assign conflicting predictions to individual data points during test time. In this work, we study this sensitivity in recommender systems, where users' recommendations are drastically altered by minor perturbations in other unrelated users' interactions. We introduce a measure of stability for recommender systems, called Rank List Sensitivity (RLS), which measures how rank lists generated by a given recommender system at test time change as a result of a perturbation in the training data. We develop a method, CASPER, which uses cascading effect to identify the minimal and systematical perturbation to induce higher instability in a recommender system. Experiments on four datasets show that recommender models are overly sensitive to minor perturbations introduced randomly or via CASPER - even perturbing one random interaction of one user drastically changes the recommendation lists of all users. Importantly, with CASPER perturbation, the models generate more unstable recommendations for low-accuracy users (i.e., those who receive low-quality recommendations) than high-accuracy ones.