In SAS Model Manager, what is the purpose of designating a model as the 'champion' model in a champion/challenger deployment pattern?
Select an answer to reveal the explanation.
Short Explanation
Here's the deal — the champion model is the one that gets the bulk of live production scoring traffic, while challenger models get a smaller slice for evaluation. You compare their performance to decide whether to promote a challenger.
Full Explanation
The correct choice is the one stating that the champion model receives the majority (or all) of production scoring traffic while challenger models receive a smaller portion to evaluate their performance. In the champion/challenger pattern, the champion handles the bulk of live production traffic while one or more challenger models are exposed to a controlled fraction of traffic. Performance metrics from both are compared to decide whether to promote a challenger. The automatic nightly retraining option is wrong because automatic retraining is a separate capability. The archived-and-replaced option describes what happens when a challenger wins — the champion is replaced — but this is the outcome, not the definition of champion status. The highest training accuracy option is wrong because champion status is based on production deployment, not training accuracy.