Glossary term
Glossary term
Foundations
In machine learning, a situation in which a model's predictions influence the training data for the same model or another model. For example, a model that recommends movies will influence the movies that people see, which will then influence subsequent movie recommendation models.
See Production ML systems: Questions to ask in Machine Learning Crash Course for more information.
Created for this library
A retention team is careful to break the feedback loop where the model targets only customers who already received outreach, masking true churn drivers.
An ad-platform team monitors the feedback loop between the model's ranking and observed clicks to prevent reinforcement of early biases.
A search-quality team uses interleaving experiments to break the feedback loop where the production ranker biases the training data it later sees.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License