AWS · MLA-C02 · BETA
Overfitting, ensembles, and model explanations
MLA-C02 · Lesson 10 of 23 · 1:15:17
YouTube loads only after you press play. This chapter opens at its timestamp in the full course. Watch on YouTube ↗
In this lesson
Lesson 10 of the free MLA-C02 beta series (AWS Certified Machine Learning Engineer – Associate, domain 2, model development). In this lesson: overfitting, underfitting, and catastrophic forgetting, ensembles (bagging, boosting, stacking, cascades), training that does not converge, and explanations with SageMaker Clarify and SHAP values. With three quick checks and a recap, and AnyCompany Mobility, a car sharing service, as the running example.
Preview the topics
- The deposit model reaches 99% accuracy on the training data and 71% on the validation data. What should the ML engineer do?
- A deep network trains normally for a few hundred steps, then its gradients grow larger at every step, and the loss turns into NaN. The inputs are already normalized. What should the ML engineer do?
- A support agent must explain to one customer which features led the model to ask for a higher deposit. What should the ML engineer give the agent?