AWS Certified AI Practitioner Practice Exam

Which of the following is essential to increase when training a model that does not perform well in production?

1. Decrease the complexity of the model

2. Increase the volume of training data

3. Optimize hyperparameters rigorously

4. Extend the training duration significantly

The correct answer is: Increase the volume of training data

Increasing the volume of training data is essential when improving the performance of a model that struggles in production. A common reason for poor model performance is that it lacks sufficient data to learn the underlying patterns effectively. By augmenting the training dataset, you provide the model with more diverse examples, enabling it to generalize better to unseen data and improve its predictive capability. Moreover, a larger dataset helps to mitigate issues related to overfitting, where a model learns to memorize the training data rather than generalize from it. Additional data can lead to a more robust model that performs well across various conditions, thereby enhancing its effectiveness in real-world applications. This answer highlights the importance of data quantity in machine learning, particularly for models facing performance challenges.