MLS-C01 · Question #52
A Machine Learning Specialist kicks off a hyperparameter tuning job for a tree-based ensemble model using Amazon SageMaker with Area Under the ROC Curve (AUC) as the objective metric. This workflow…
The correct answer is D. A scatter plot showing the correlation between maximum tree depth and the objective metric. The goal is to reconfigure hyperparameter ranges to reduce training time and cost. To do this intelligently, the Specialist needs to see which values of a hyperparameter correlate with high objective metric (AUC) performance - allowing them to narrow the search space to the…
Question
A Machine Learning Specialist kicks off a hyperparameter tuning job for a tree-based ensemble model using Amazon SageMaker with Area Under the ROC Curve (AUC) as the objective metric. This workflow will eventually be deployed in a pipeline that retrains and tunes hyperparameters each night to model click-through on data that goes stale every 24 hours. With the goal of decreasing the amount of time it takes to train these models, and ultimately to decrease costs, the Specialist wants to reconfigure the input hyperparameter range(s). Which visualization will accomplish this?
Options
- AA histogram showing whether the most important input feature is Gaussian.
- BA scatter plot with points colored by target variable that uses t-Distributed Stochastic Neighbor
- CA scatter plot showing the performance of the objective metric over each training iteration.
- DA scatter plot showing the correlation between maximum tree depth and the objective metric.
How the community answered
(68 responses)- A3% (2)
- B7% (5)
- C13% (9)
- D76% (52)
Explanation
The goal is to reconfigure hyperparameter ranges to reduce training time and cost. To do this intelligently, the Specialist needs to see which values of a hyperparameter correlate with high objective metric (AUC) performance - allowing them to narrow the search space to the promising region. A scatter plot of maximum tree depth vs. AUC (D) directly shows this correlation: if good AUC is achieved only within a specific depth range, the Specialist can set tighter bounds, reducing the number of training jobs needed. A histogram of feature distributions (A) reveals data properties, not hyperparameter behavior. A t-SNE scatter plot (B) is a dimensionality reduction visualization technique unrelated to hyperparameter tuning insight. A scatter plot of objective metric over training iterations (C) shows convergence behavior but does not reveal which hyperparameter values drive performance, offering no guidance on how to narrow the search range.
Topics
Community Discussion
No community discussion yet for this question.