25CONFIDENCE INTERVALS
What are the different ways to construct confidence intervals for machine learning classifiers?
There are several ways to construct confidence intervals for machine learning models, depending on the model type and the nature of your data. For instance, some methods are computationally expensive when working with deep neural networks and are thus more suitable to less resource-intensive machine learning models. Others require larger datasets to be reliable.
The following are the most common methods for constructing confidence intervals:
- Constructing normal approximation intervals based on a test set
- Bootstrapping training ...
Get Machine Learning Q and AI now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
