Existence conditions for hidden feedback loops in online recommender systems
We explore a hidden feedback loops effect in online recommender systems. Feedback loops result in degradation of online multi-armed bandit (MAB) recommendations to a small subset and loss of coverage and novelty. We study how uncertainty and noise in user interests influence the existence of feedback loops. First, we show that an unbiased additive random noise in user interests does not prevent a feedback loop. Second, we demonstrate that a non-zero probability of resetting user interests is sufficient to limit the feedback loop and estimate the size of the effect. Our experiments confirm the theoretical findings in a simulated environment for four bandit algorithms.
Code (1)
Tasks
Recommendation SystemsSimilar Papers 제목 키워드 기반
Analysis of hidden feedback loops in continuous machine learning systems
In this concept paper, we discuss intricacies of specifying and verifying the quality of continuous and lifelong learning artificial intelligence systems as they interact with and influence their environment causing a so…
BIG-bench Machine LearningLifelong learningInherent directionality explains the lack of feedback loops in empirical networks
We explore the hypothesis that the relative abundance of feedback loops in many empirical complex networks is severely reduced owing to the presence of an inherent global directionality. Aimed at quantifying this idea, w…
DPR: An Algorithm Mitigate Bias Accumulation in Recommendation feedback loops
Recommendation models trained on the user feedback collected from deployed recommendation systems are commonly biased. User feedback is considerably affected by the exposure mechanism, as users only provide feedback on t…
Recommendation SystemsA Classification of Feedback Loops and Their Relation to Biases in Automated Decision-Making Systems
Prediction-based decision-making systems are becoming increasingly prevalent in various domains. Previous studies have demonstrated that such systems are vulnerable to runaway feedback loops, e.g., when police are repeat…
Decision MakingFairnessRecommendation SystemsRelationPeer Effect Estimation in the Presence of Simultaneous Feedback and Unobserved Confounders
Estimating peer causal effects within complex real-world networks such as social networks is challenging, primarily due to simultaneous feedback between peers and unobserved confounders. Existing methods either address u…