AI engineering glossary · Time series recsys
What is IPS (inverse propensity scoring)?
Inverse propensity scoring (also called IPS): A correction method that weights each logged observation by one divided by the probability that it was shown or chosen. It gives an unbiased estimate of performance under a different policy, but can have high variance when probabilities are tiny.
Inverse propensity scoring: an example
A click on an item shown with probability 0.1 counts as 10 clicks in the estimate.
Related terms
More Time series recsys terms
- Item-based collaborative filtering
- Kalman filter
- KPSS test
- LambdaMART
- Latent factors
- Learning to rank
- Listwise ranking
- Ljung-Box test
Where you learn it
Inverse propensity scoring is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.