Lesson 4
20 min

Training Data, Loss, and Gradient Descent (Plain English)

Quick Summary

Training data supplies examples, a loss function measures error, and gradient descent guides small weight updates. Together they define what the model is rewarded for learning.

What you will learn
  • ·Explain the roles of training, validation, and test data
  • ·Describe loss as a measurement of model error
  • ·Use a plain-language analogy for gradient descent