Skip to content
AI.info

AI.info

Recommender systems

Decision pipelines that shape what millions of people see, and the objectives that quietly decide for them.

  1. Recommender Systems as Decision Pipelines
  2. Product Surfaces, Objectives, and Non-Goals
  3. Users, Items, Context, and Recommendation Units
  4. Interaction Logs, Exposure, and Missing-Not-at-Random Data
  5. Explicit, Implicit, and Negative Feedback
  6. Labels, Rewards, Delays, and Proxy Outcomes
  7. Eligibility, Inventory, and Candidate Constraints
  8. Baselines, Heuristics, and Popularity Recommenders
  9. User-Based Collaborative Filtering
  10. Item-Based Collaborative Filtering
  11. Similarity, Normalization, Shrinkage, and Sparsity
  12. Matrix Factorization for Explicit Feedback
  13. Implicit Matrix Factorization and Confidence Weighting
  14. Bayesian Personalized Ranking and Pairwise Learning
  15. Content-Based Recommendation
  16. Hybrid Recommendation and Cold-Start Strategy
  17. Embeddings and Two-Tower Retrieval
  18. Negative Sampling and Retrieval Training
  19. Approximate Nearest-Neighbor Search and Vector Indexes
  20. Candidate Generation as a Portfolio
  21. Graph-Based Recommendation and Random Walks
  22. Graph Neural Recommenders and LightGCN
  23. Taxonomies, Knowledge Graphs, and Structured Item Relations
  24. Session-Based Recommendation
  25. Sequential Recommendation and Next-Item Prediction
  26. Temporal Dynamics, Recency, and Seasonality
  27. Context-Aware Recommendation
  28. User Modeling: Long-Term Taste and Short-Term Intent
  29. Multimodal Item Understanding
  30. Deep Ranking Models and Feature Interactions
  31. Pointwise, Pairwise, and Listwise Ranking Objectives
  32. Multi-Stage Ranking and Feature Serving
  33. Multi-Objective Ranking and Constraint Handling
  34. Re-Ranking for Diversity, Novelty, Serendipity, and Coverage
  35. Calibration and Preference Consistency
  36. Slates, Page Layout, and Position Interactions
  37. Exploration, Contextual Bandits, and Online Learning
  38. Offline Evaluation Protocols and Temporal Splits
  39. Ranking Metrics and Metric Portfolios
  40. Counterfactual Evaluation, Propensity, and Position Bias
  41. Online Experiments, Guardrails, and Long-Term Value
  42. Causal Questions, Incrementality, and Policy Effects
  43. Multi-Stakeholder and Marketplace Recommendation
  44. Feedback Loops, Popularity Bias, and Ecosystem Dynamics
  45. Fairness, Exposure, and User Control
  46. Privacy, Personalization Boundaries, and Data Governance
  47. Production Architecture, Freshness, Monitoring, and Incidents
  48. Conversational, Generative, and Foundation-Model Recommenders
  49. Recommender Lifecycle, Change Management, and Retirement
  50. Capstone: Design and Defend a Recommendation System