Learning Disentangled Representations for Recommendation

Abstract

User behavior data in recommender systems are driven by the complex interactions of many latent factors behind the users' decision making processes. The factors are highly entangled, and may range from high-level ones that govern user intentions, to low-level ones that characterize a user's preference when executing an intention. Learning representations that uncover and disentangle these latent factors can bring enhanced robustness, interpretability, and controllability. However, learning such disentangled representations from user behavior is challenging, and remains largely neglected by the existing literature. In this paper, we present the MACRo-mIcro Disentangled Variational Auto-Encoder (MacridVAE) for learning disentangled representations from user behavior. Our approach achieves macro disentanglement by inferring the high-level concepts associated with user intentions (e.g., to buy a shirt or a cellphone), while capturing the preference of a user regarding the different concepts separately. A micro-disentanglement regularizer, stemming from an information-theoretic interpretation of VAEs, then forces each dimension of the representations to independently reflect an isolated low-level factor (e.g., the size or the color of a shirt). Empirical results show that our approach can achieve substantial improvement over the state-of-the-art baselines. We further demonstrate that the learned representations are interpretable and controllable, which can potentially lead to a new paradigm for recommendation where users are given fine-grained control over targeted aspects of the recommendation lists.

To appear in the Proceedings of the Thirty-third Conference on Neural Information Processing Systems (NeurIPS 2019)

Cited by in corpus (34)

• Learning Intents behind Interactions with Knowledge Graph for Recommendation
• Advances and Challenges in Conversational Recommender Systems: A Survey
• DisenHAN: Disentangled Heterogeneous Graph Attention Network for Recommendation
• Deep Learning on Graphs: A Survey
• Disentangling Hate in Online Memes
• Factorizable Graph Convolutional Networks
• Automated Machine Learning on Graphs: A Survey
• RecBole: Towards a Unified, Comprehensive and Efficient Framework for Recommendation Algorithms
• SimpleX: A Simple and Strong Baseline for Collaborative Filtering
• Robust Federated Recommendation System
• CausalVAE: Structured Causal Disentanglement in Variational Autoencoder
• Disentangling User Interest and Conformity for Recommendation with Causal Embedding
• Controllable Multi-Interest Framework for Recommendation
• Intent Disentanglement and Feature Self-supervision for Novel Recommendation
• Popularity Bias Is Not Always Evil: Disentangling Benign and Harmful Bias for Recommendation
• Contrastive Learning for Debiased Candidate Generation in Large-Scale Recommender Systems
• From Intrinsic to Counterfactual: On the Explainability of Contextualized Recommender Systems
• Cross-modal Variational Auto-encoder for Content-based Micro-video Background Music Recommendation
• Adversarial Attacks and Defenses: An Interpretation Perspective
• Accurate Bundle Matching and Generation via Multitask Learning with Partially Shared Parameters
• Adversarial and Contrastive Variational Autoencoder for Sequential Recommendation
• Task-adaptive Neural Process for User Cold-Start Recommendation
• Information Diffusion Prediction with Latent Factor Disentanglement
• Edge-Enhanced Global Disentangled Graph Neural Network for Sequential Recommendation
• Click-through Rate Prediction with Auto-Quantized Contrastive Learning
• Explainable Recommender Systems via Resolving Learning Representations
• Predicting the Accuracy of a Few-Shot Classifier
• Hierarchical User Intent Graph Network forMultimedia Recommendation
• Semi-Disentangled Representation Learning in Recommendation System
• Controllable Gradient Item Retrieval
• Intrinsic and Extrinsic Factor Disentanglement for Recommendation in Various Context Scenarios
• Controllable Recommenders using Deep Generative Models and Disentanglement
• Graphs as Tools to Improve Deep Learning Methods
• Sparse-Interest Network for Sequential Recommendation