About this document
Machine Learning Model Evaluation Methods by Yen-Kai Cheng is a document available to read on EtoBox.
The document discusses how machine learning models are optimized by evaluating them on a metric like error on the target population and aiming to find the model with the lowest error. It introduces the concept of an error surface that maps models to their error, and how gradient descent can be used as an optimization method to iteratively adjust model parameters downhill following the steepest descent of the error surface to locate an optimal model. However, gradient descent may find local optima instead of
- Author
- Yen-Kai Cheng
- Language
- EN