Track 11: ML and Deep Learning Foundations for AI Engineers, Maths Intuition

Gradients, Loss Functions and Gradient Descent

Explain how a model learns by going downhill.

Status: Coming soon IntermediateVideo lessonAbout 12 min Free tools

Part of Train and Fine-Tune Models (3 of 45)

Get notified on YouTube Udemy certificate: coming soon

Course

Transformers & GPT From Scratch for AI Engineers

Udemy: coming soon

Tools

  • NumPy (open-source, free)

Watch first

Then try

In this module

  1. Vectors, Similarity and Probability, Visually
  2. Matrices: What a Layer Really Does
  3. Gradients, Loss Functions and Gradient Descent
  4. Optimisers: SGD, Adam, Learning-Rate Schedules

The course this lesson belongs to