A Meta-Learning Perspective on Cold-Start Recommendations for Items
Explore this paper's citation graph
Summary
This paper proposes two deep neural network architectures that implement a meta-learning strategy to address item cold-start when new items arrive continuously and demonstrates that these techniques significantly beat the MF baseline and also outperform production models for Tweet recommendation.
- Type
- article
- Published
- 2017-01-01
- Cited by
- 233
- References
- 29
- OpenAlex
- https://openalex.org/W2753433947
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:45426075
Keywords
Computer science, Cold start (automotive), Deep learning, Artificial intelligence, Machine learning
References
- Industry Report: Amazon.com Recommendations: Item-to-Item Collaborative Filtering
- Leveraging user libraries to bootstrap collaborative filtering
- Factorization meets the neighborhood: a multifaceted collaborative filtering model
- Addressing cold-start problem in recommendation systems
- Recurrent neural networks
- Regression-based latent factor models
- Practical Lessons from Predicting Clicks on Ads at Facebook
- Metalearning: a survey of trends and technologies
- Google news personalization: scalable online collaborative filtering
- Matchbox: large scale online bayesian recommendations
- Collaborative topic modeling for recommending scientific articles
- Probabilistic Matrix Factorization
- Co-factorization machines: modeling user interests and predicting individual decisions in Twitter
- A Perspective View and Survey of Meta-Learning
- Personalized news recommendation based on click behavior
- Human-level concept learning through probabilistic program induction
- Parallel Recurrent Neural Network Architectures for Feature-rich Session-based Recommendations
- Deep Neural Networks for YouTube Recommendations
- Convolutional Matrix Factorization for Document Context-Aware Recommendation
- Meta-Learning with Memory-Augmented Neural Networks
Cited by
- Contextual Explanation Networks
- Deep Learning Based Recommender System
- Latent Cross: Making Use of Context in Recurrent Recommender Systems
- Federated Meta-Learning for Recommendation
- Neural variational entity set expansion for automatically populated knowledge graphs
- Meta-Learning: A Survey
- Automatic Music Playlist Continuation via Neighbor-based Collaborative Filtering and Discriminative Reweighting/Reranking
- Large-scale Collaborative Filtering with Product Embeddings
- Collaborative Sensing with Interactive Learning using Dynamic Intelligent Virtual Sensors
- New Item Consumption Prediction Using Deep Learning
- A Content-Based Approach to Email Triage Action Prediction: Exploration and Evaluation
- Adaptive Deep Modeling of Users and Items Using Side Information for Recommendation
- Representation Learning for Words and Entities
- Fast and Flexible Multi-Task Classification Using Conditional Neural Adaptive Processes
- On Both Cold-Start and Long-Tail Recommendation with Social Data
- Warm Up Cold-start Advertisements: Improving CTR Predictions via Learning to Learn ID Embeddings
- Sequential Scenario-Specific Meta Learner for Online Recommendation
- MeLU: Meta-Learned User Preference Estimator for Cold-Start Recommendation
- RecoNet: An Interpretable Neural Architecture for Recommender Systems
- Time-weighted Attentional Session-Aware Recommender System
Related papers
- Prototypical Networks for Few-shot Learning
- Neural Collaborative Filtering
- Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
- Matrix Factorization Techniques for Recommender Systems
- MeLU: Meta-Learned User Preference Estimator for Cold-Start Recommendation
- Optimization as a Model for Few-Shot Learning
- Factorization Machines
- Wide & Deep Learning for Recommender Systems