Answer: It adds class weights and downsamples the major class(es).

AutoML employs strategies to manage imbalanced datasets, where one class significantly outnumbers others. It assigns higher weights to less frequent classes during training to highlight their importance and prevent bias towards the majority class. Additionally, it may downsample the major classes to achieve a more balanced data distribution, enhancing the model's ability to learn from both minority and majority classes. This approach is vital for improving model performance and preventing bias, ensuring accurate predictions for minority classes, which are often critical. Unlike other options, AutoML does not discard the dataset or upsample the minority class as its primary strategy, nor does it alter the class distributions in test and validation sets, maintaining their integrity for accurate evaluation.

Author: LeetQuiz Editorial Team