Learning to Warm Up Cold Item Embeddings for Cold-start Recommendation with Meta Scaling and Shifting Networks

Abstract

Recently, embedding techniques have achieved impressive success in recommender systems. However, the embedding techniques are data demanding and suffer from the cold-start problem. Especially, for the cold-start item which only has limited interactions, it is hard to train a reasonable item ID embedding, called cold ID embedding, which is a major challenge for the embedding techniques. The cold item ID embedding has two main problems: (1) A gap is existing between the cold ID embedding and the deep model. (2) Cold ID embedding would be seriously affected by noisy interaction. However, most existing methods do not consider both two issues in the cold-start problem, simultaneously. To address these problems, we adopt two key ideas: (1) Speed up the model fitting for the cold item ID embedding (fast adaptation). (2) Alleviate the influence of noise. Along this line, we propose Meta Scaling and Shifting Networks to generate scaling and shifting functions for each item, respectively. The scaling function can directly transform cold item ID embeddings into warm feature space which can fit the model better, and the shifting function is able to produce stable embeddings from the noisy embeddings. With the two meta networks, we propose Meta Warm Up Framework (MWUF) which learns to warm up cold ID embeddings. Moreover, MWUF is a general framework that can be applied upon various existing deep recommendation models. The proposed model is evaluated on three popular benchmarks, including both recommendation and advertising datasets. The evaluation results demonstrate its superior performance and compatibility.

arXiv comment10 pages, accepted by SIGIR2021

References in corpus (20)

• Adam: A Method for Stochastic Optimization
• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• Prototypical Networks for Few-shot Learning
• BPR: Bayesian Personalized Ranking from Implicit Feedback
• RippleNet: Propagating User Preferences on the Knowledge Graph for Recommender Systems
• xDeepFM: Combining Explicit and Implicit Feature Interactions for Recommender Systems
• Sequential Recommender Systems: Challenges, Progress and Prospects
• Neural Factorization Machines for Sparse Predictive Analytics
• ESAM: Discriminative Domain Adaptation with Non-Displayed Items to Improve Long-Tail Performance
• Modeling Users' Behavior Sequences with Hierarchical Explainable Network for Cross-domain Fraud Detection
• How to Retrain Recommender System? A Sequential Meta-Learning Method
• MeLU: Meta-Learned User Preference Estimator for Cold-Start Recommendation
• Product-based Neural Networks for User Response Prediction
• LGM-Net: Learning to Generate Matching Networks for Few-Shot Learning
• Real-time Attention Based Look-alike Model for Recommender System
• Warm Up Cold-start Advertisements: Improving CTR Predictions via Learning to Learn ID Embeddings
• MAMO: Memory-Augmented Meta-Optimization for Cold-start Recommendation
• Modelling of Bi-directional Spatio-Temporal Dependence and Users' Dynamic Preferences for Missing POI Check-in Identification
• Modeling the Field Value Variations and Field Interactions Simultaneously for Fraud Detection
• Neural Hierarchical Factorization Machines for User's Event Sequence Analysis

Cited by in corpus (14)

• LightFR: Lightweight Federated Recommendation with Privacy-preserving Matrix Factorization
• Contrastive Collaborative Filtering for Cold-Start Item Recommendation
• User-Centric Conversational Recommendation with Multi-Aspect User Modeling
• Improving Item Cold-start Recommendation via Model-agnostic Conditional Variational Autoencoder
• Domain Disentanglement with Interpolative Data Augmentation for Dual-Target Cross-Domain Recommendation
• Online Item Cold-Start Recommendation with Popularity-Aware Meta-Learning
• Proxy-based Item Representation for Attribute and Context-aware Recommendation
• Optimizing E-commerce Search: Toward a Generalizable and Rank-Consistent Pre-Ranking Model
• AutoAssign+: Automatic Shared Embedding Assignment in Streaming Recommendation
• POSO: Personalized Cold Start Modules for Large-scale Recommender Systems
• Click-through Rate Prediction with Auto-Quantized Contrastive Learning
• VM-Rec: A Variational Mapping Approach for Cold-start User Recommendation
• Harnessing Light for Cold-Start Recommendations: Leveraging Epistemic Uncertainty to Enhance Performance in User-Item Interactions
• Scene-wise Adaptive Network for Dynamic Cold-start Scenes Optimization in CTR Prediction