AI engineering glossary · Time series recsys
What is BPR (bayesian personalised ranking)?
Bayesian personalised ranking (also called BPR): A training approach for implicit feedback that learns to score items a user interacted with above items they did not, by optimising pairwise preferences. Formula per pair: maximise log sigmoid(score(user, positive) - score(user, negative)).
Bayesian personalised ranking: an example
For a user who bought item A but not B, the model is trained so that score(A) exceeds score(B).
Related terms
More Time series recsys terms
- Box-Jenkins method
- Calendar features
- Candidate generation
- Changepoint
- Click-through rate
- Click-through rate prediction
- Cointegration
- Collaborative filtering
Where you learn it
Bayesian personalised ranking is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.