AI engineering glossary · Time series recsys
What is candidate generation?
Candidate generation (also called Retrieval stage): The first stage of a large recommender, which quickly narrows millions of items to a few hundred plausible ones using cheap methods such as embeddings and nearest-neighbour search. Recall matters here, since anything dropped cannot be shown.
Candidate generation: an example
From 10 million videos, fetch the 500 whose embeddings are closest to the user's embedding.
Related terms
More Time series recsys terms
- Changepoint
- Click-through rate
- Click-through rate prediction
- Cointegration
- Collaborative filtering
- Content-based filtering
- Conversion rate
- Correlogram
Where you learn it
Candidate generation is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.