Robust Recommendation from Noisy Implicit Feedback: A GMM-Weighted Bayes-label Transition Matrix Framework

2026-05-20

AI Summary

This work addresses the pervasive issue of label noise in implicit-feedback recommendation systems, where existing denoising approaches suffer from either low data utilization due to sample discarding or performance degradation caused by biased estimation of Bayesian label transition matrices. To overcome these limitations, the authors propose RGBT, a novel framework that, for the first time, integrates Gaussian Mixture Models (GMMs) to compute a reliability score for each sample, enabling sample-level weighted calibration. This strategy retains all training samples while yielding a consistent and low-variance unbiased estimate of the transition matrix. Extensive experiments demonstrate that RGBT significantly outperforms current state-of-the-art methods—both reliable-sample selection and transition-matrix-based approaches—across multiple real-world and synthetically corrupted datasets, achieving superior performance in both noise exploitation and calibration accuracy.

Abstract

Learning from implicit feedback in recommender systems is fundamentally challenged by pervasive label noise. While conventional denoising approaches often discard noisy instances to ensure robustness, this strategy inevitably suffers from low data utilization. Alternative methods that employ a Bayes-label transition matrix (BLTM) can leverage all available data, but their estimates tend to be biased in practical recommendation scenarios. To address these limitations, this paper proposes a Robust GMM-weighted Bayes-label Transition Matrix framework (RGBT). Our solution utilizes a Gaussian Mixture Model (GMM) to derive instance-specific reliability scores, which systematically calibrate the BLTM estimation to mitigate bias. Theoretical analysis confirms that our approach, by leveraging the BLTM framework with GMM calibration, simultaneously ensures full sample utilization, delivers consistent estimation, and critically, achieves a significant reduction in estimation variance. Extensive experiments on multiple real-world and synthetically flipped datasets demonstrate that RGBT not only utilizes noisy samples more effectively than mainstream reliable sample-based denoising methods, but also achieves significantly superior calibration capability of the transition matrix compared to state-of-the-art transition matrix-based denoising approaches.

Problem

• implicit feedback
• label noise
• recommender systems
• Bayes-label transition matrix
• data utilization

Innovation

• Gaussian Mixture Model
• Bayes-label Transition Matrix
• Implicit Feedback
• Label Noise
• Robust Recommendation