Doubly Calibrated Estimator for Recommendation on Data Missing Not At Random

arXiv:2403.00817v1 Announce Type: cross
Abstract: Recommender systems often suffer from selection bias as users tend to rate their preferred items. The datasets collected under such conditions exhibit entries missing not at random and thus are not randomized-controlled trials representing the target population. To address this challenge, a doubly robust estimator and its enhanced variants have been proposed as they ensure unbiasedness when accurate imputed errors or predicted propensities are provided. However, we argue that existing estimators rely on miscalibrated imputed …

abstract arxiv bias challenge cs.ir cs.lg data datasets population random rate recommendation recommender systems robust systems type