Why People Skip Music? On Predicting Music Skips using Deep Reinforcement Learning

Abstract

Music recommender systems are an integral part of our daily life. Recent research has seen a significant effort around black-box recommender based approaches such as Deep Reinforcement Learning (DRL). These advances have led, together with the increasing concerns around users' data collection and privacy, to a strong interest in building responsible recommender systems. A key element of a successful music recommender system is modelling how users interact with streamed content. By first understanding these interactions, insights can be drawn to enable the construction of more transparent and responsible systems. An example of these interactions is skipping behaviour, a signal that can measure users' satisfaction, dissatisfaction, or lack of interest. In this paper, we study the utility of users' historical data for the task of sequentially predicting users' skipping behaviour. To this end, we adapt DRL for this classification task, followed by a post-hoc explainability (SHAP) and ablation analysis of the input state representation. Experimental results from a real-world music streaming dataset (Spotify) demonstrate the effectiveness of our approach in this task by outperforming state-of-the-art models. A comprehensive analysis of our approach and of users' historical data reveals a temporal data leakage problem in the dataset. Our findings indicate that, overall, users' behaviour features are the most discriminative in how our proposed DRL model predicts music skips. Content and contextual features have a lesser effect. This suggests that a limited amount of user data should be collected and leveraged to predict skipping behaviour.

arXiv commentAccepted at CHIIR 2023

References in corpus (19)

• Addressing Function Approximation Error in Actor-Critic Methods
• Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problems
• Deep Recurrent Q-Learning for Partially Observable MDPs
• Deep Reinforcement Learning for Page-wise Recommendations
• Stabilizing Off-Policy Q-Learning via Bootstrapping Error Reduction
• Denoising Implicit Feedback for Recommendation
• Off-Policy Deep Reinforcement Learning without Exploration
• Reinforcement Learning for Relation Classification from Noisy Data
• Benchmarking Batch Deep Reinforcement Learning Algorithms
• An Optimistic Perspective on Offline Reinforcement Learning
• Deep Reinforcement Learning based Recommendation with Explicit User-Item Interactions Modeling
• Counterfactual Evaluation of Slate Recommendations with Sequential Reward Interactions
• Making Neural Networks Interpretable with Attribution: Application to Implicit Signals Prediction
• Classification with Costly Features as a Sequential Decision-Making Problem
• The Skipping Behavior of Users of Music Streaming Services and its Relation to Musical Structure
• Datum-Wise Classification: A Sequential Approach to Sparsity
• Sequential Skip Prediction with Few-shot in Streamed Music Contents
• Session-based Sequential Skip Prediction via Recurrent Neural Networks
• Sequential modeling of Sessions using Recurrent Neural Networks for Skip Prediction

Cited by in corpus (2)

• Learning and Optimization of Implicit Negative Feedback for Industrial Short-video Recommender System
• Negative Feedback for Music Personalization