Practical ML

• Problem Framing
  • Multi-Label Learning
  • Multi-Task Learning
• Understanding Raw Data
  • User
  • Item
  • Interactions
  • Search and Recommender Systems
• Labeling and Learning Strategies
  • Resources
  • Label Design
  • Data and Feature Engineering
  • Dataset Creation and Curation
• Model
  • Resources
  • Retrieval
  • Ranking
• Evaluation
  • Offline Evaluation
  • Online Evaluation
  • LLM App Evaluation
• Applied Causal Inference & Uplift Modeling
• Retraining
• Practical Resources
• Review Topics
• Problems: Bias
  • Cheat Sheet
  • Feedback Loop
  • Label Leakage
  • Freshness and Fatigue Bias
• MLSD Problems
  • Section 1: Modeling Paradigm and Loss Selection
  • Section 2: Evaluation Metric Alignment
  • Section 3: Joint and Multi-Task Training
  • Section 4: Architecture and Scalability Trade-offs
  • Section 5: Debugging and Failure Mode Diagnosis
  • Section 6: Cross-Modal Retrieval, Ranking, and Personalization
  • Section 7: Latency Constraints and Inference Optimizations
  • Section 8: ANN-Specific Retrieval Challenges
  • Section 9: Tail Query Recovery and Head Bias
  • Section 10: Product Categorization in Marketplace
  • Section 11: Ads Moderation: Modeling + System Design
  • Section 12: Content Understanding: Taxonomy + Semantics
  • Section 13: Difficult Data Regimes