Abstract:With the popularity of online music services, personalized music recommendation has garnered much research interest. Recommendation models are typically trained on datasets constructed from user feedback, which includes both the active feedback (e.g., clicking the Like or Skip buttons) and passive feedback (e.g., auto-play), with passive feedback comprising the majority. Due to the unavailability of user attention, the massive amount of passive feedback is unreliable, significantly compromising the quality of the training data. How to estimate the user's attention on the target music has become a critical problem in music recommendation. Heuristic methods such as exponential decay and negative sampling have been proposed. However, they either neglect the sequential dependencies between feedback actions or utilize only a small fraction of passive samples, leading to inaccurate and biased attention estimation. In this paper, we naturally propose modeling user attention prediction as a positive-unlabeled (PU) learning problem, where active feedback is treated as positive samples and passive feedback is treated as unlabeled samples, as we can only ensure that the user's attention is focused when she provides active feedback. Then we propose an extended PU-learning model with sequential dependencies, called UAE, which contains an unbiased user attention estimator and an unbiased propensity estimator. Subsequently, a joint learning algorithm is developed in which the attention and propensity estimators are optimized in alternating fashion. Theoretical analysis shows the unbiasedness and variance of the attention estimator and the propensity estimator. Extensive experiments on two large-scale datasets demonstrate the proposed UAE's effectiveness and generality in enhancing downstream music recommendation. One week online A/B testing on Huawei Music App manifests that UAE can significantly increase the users' play count and time over 2%, further demonstrating the effectiveness of UAE in real-world music recommendation products.

Exploiting Negative Preference in Content-based Music Recommendation with Contrastive Learning

Recommending More Suitable Music Based on Users' Real Context.

Leveraging Negative Signals with Self-Attention for Sequential Music Recommendation

Exploiting the emotional preference of music for music recommendation in daily activities

Enhancing Sequential Music Recommendation with Negative Feedback-informed Contrastive Learning

Negative Feedback for Music Personalization

Contrastive Learning with Positive-Negative Frame Mask for Music Representation

Leveraging the structure of musical preference in content-aware music recommendation

Loss Aversion in Recommender Systems: Utilizing Negative User Preference to Improve Recommendation Quality

Learning to Embed Music and Metadata for Context-Aware Music Recommendation

Learning content similarity for music recommendation

Improving Music Performance Assessment with Contrastive Learning

Modeling User Attention in Music Recommendation

Personalized Representation with Contrastive Loss for Recommendation Systems

Learning Recommendation Algorithm Based on Improved BP Neural Network in Music Marketing Strategy

Deep Content-User Embedding Model for Music Recommendation

Negative Feedback Aware Hybrid Sequential Neural Recommendation Model.

Contrastive Learning with Reverse Extended Enhancement for Sequential Recommendation

Contrastive Learning for Recommender System

Cracking the Code of Negative Transfer: A Cooperative Game Theoretic Approach for Cross-Domain Sequential Recommendation

A Gain-Tuning Dynamic Negative Sampler for Recommendation