MeLU: Meta-Learned User Preference Estimator for Cold-Start Recommendation

Abstract

This paper proposes a recommender system to alleviate the cold-start problem that can estimate user preferences based on only a small number of items. To identify a user's preference in the cold state, existing recommender systems, such as Netflix, initially provide items to a user; we call those items evidence candidates. Recommendations are then made based on the items selected by the user. Previous recommendation studies have two limitations: (1) the users who consumed a few items have poor recommendations and (2) inadequate evidence candidates are used to identify user preferences. We propose a meta-learning-based recommender system called MeLU to overcome these two limitations. From meta-learning, which can rapidly adopt new task with a few examples, MeLU can estimate new user's preferences with a few consumed items. In addition, we provide an evidence candidate selection strategy that determines distinguishing items for customized preference estimation. We validate MeLU with two benchmark datasets, and the proposed model reduces at least 5.92% mean absolute error than two comparative models on the datasets. We also conduct a user study experiment to verify the evidence selection strategy.

Accepted as a full paper at KDD 2019

References in corpus (4)

• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• Prototypical Networks for Few-shot Learning
• Meta-SGD: Learning to Learn Quickly for Few-Shot Learning
• Memory Matching Networks for One-Shot Image Recognition

Cited by in corpus (17)

• Learning to Warm Up Cold Item Embeddings for Cold-start Recommendation with Meta Scaling and Shifting Networks
• Contrastive Collaborative Filtering for Cold-Start Item Recommendation
• Improving Item Cold-start Recommendation via Model-agnostic Conditional Variational Autoencoder
• Cross-Market Product Recommendation
• CDRNP: Cross-Domain Recommendation to Cold-Start Users via Neural Process
• Doing More with Less: Overcoming Data Scarcity for POI Recommendation via Cross-Region Transfer
• INMO: A Model-Agnostic and Scalable Module for Inductive Collaborative Filtering
• Meta-Learning with Adaptive Weighted Loss for Imbalanced Cold-Start Recommendation
• A Machine Learning Approach for Automated Filling of Categorical Fields in Data Entry Forms
• Automating Outlier Detection via Meta-Learning
• AutoAssign+: Automatic Shared Embedding Assignment in Streaming Recommendation
• Adaptive-Step Graph Meta-Learner for Few-Shot Graph Classification
• Meta Learning to Rank for Sparsely Supervised Queries
• Task-adaptive Neural Process for User Cold-Start Recommendation
• Simple and Powerful Architecture for Inductive Recommendation Using Knowledge Graph Convolutions
• A Semi-Personalized System for User Cold Start Recommendation on Music Streaming Apps
• Sequential Recommendation for Cold-start Users with Meta Transitional Learning