Monitoring & Drift

Data, feature, prediction and concept drift taught separately; drift that is not failure; ground truth that arrives weeks late; and performance decay diagnosed rather than assumed.

Model Monitoring
▶ lab

A model needs everything a service needs, plus three distributions a service does not have: features in, predictions out, and outcomes back. Four layers, each with an owner, each catching a different failure.

Q · The service is up, latency is fine, and nobody knows whether the model is still right. What does a deployed model need watched, and who watches each part?
Data Drift
▶ lab

The input distribution changed. A distance metric between the training reference and this week's traffic says so on the day; whether it matters depends on where the inputs moved to, and that needs the outcomes.

Q · The inputs the model receives no longer look like the inputs it was trained on. How do you measure that, and what does the measurement license you to conclude?
Feature Drift
▶ lab

One feature's distribution moved. Before it is drift it might be a bug: a null-rate spike, a unit change, a renamed category. Diagnose the pipeline first, because retraining on a broken feature teaches the model that broken is normal.

Q · A single feature's distribution changed this week. Is that the world, or the pipeline — and what happens if you retrain before you know?
Prediction Drift
▶ lab

The output distribution moved. It is the earliest model-level signal, needs no labels, and is the one that catches train/serve skew on rollout day — because the model reacts to its inputs immediately and to the truth never.

Q · The model's score distribution changed. What does that tell you on the day, before any outcome is known, and when is it the first thing you should look at?
Concept Drift
▶ lab

The relationship between features and outcome changed. The inputs did not move, so no input monitor fires; the scores did not move, so no prediction monitor fires. Only the outcomes reveal it, and they arrive late.

Q · Every feature distribution is stable, the prediction distribution is stable, and the model is now wrong. What changed, and what could have shown it?
Drift Is Not Failure
▶ lab

A distribution can change legitimately and the model can handle it. "Drift means retrain" retrains a working model on the strength of an input metric, costs a training run and a rollout, and answers a question the metric never asked.

Q · The drift alert fired. Before anyone touches the model: does the change actually hurt, and how would you know?
Ground-Truth Delay
▶ lab

The outcome arrives weeks or months after the prediction. Every quality number on the dashboard is about the past; the architecture has to say how far past, join outcomes back by id, and use proxies honestly in the meantime.

Q · The model's quality can only be measured once the outcomes exist, and they take a quarter to arrive. What does the monitoring show for this week, and how do you avoid being reassured by last quarter?
Performance Decay
▶ lab

Quality over time went down. Five different causes produce that chart, only one of them is fixed by retraining, and two are made worse by it. Diagnose in order — bug, product change, feedback loop, data drift, concept drift — before touching the model.

Q · Model quality has been falling for a month. Which of the five things that produce that chart is it, and what happens if you retrain before you know?