Book VI
Retrieval, ranking, and recommendations
Embeddings, candidate generation, ranking, search metrics, cold start, and feedback loops.
4 chapters · 23 entriesChapters
4Retrieval foundations
Move from sparse retrieval and embeddings to approximate search and two-tower systems.
Recommendation models
Learn collaborative and feature-aware factorization methods.
Retrieval and ranking practice
Choose ranking objectives, correct biased feedback, and defend retrieval, reranking, metrics, and sampling.
Recommendation product design
Handle multi-task ranking, cold start, feedback, and ecosystem decisions.