I've been diving into machine learning through beginner-friendly YouTube videos, and I grasp the basic concepts like backpropagation, forward passing, activation functions (like softmax and ReLU), and cross-entropy loss. However, I still struggle to understand the underlying math and logic that drive these functions. For instance, I know the role of activation functions, but I'm curious about the reasoning behind their specific forms. Can anyone recommend some comprehensive sources or videos that thoroughly explain each concept in machine learning and neural networks?
1 Answer
You might want to check out the Stanford course on computer vision available on YouTube. Even though it focuses on CV, it includes important machine learning principles that are foundational to the field.
Does it cover machine learning concepts in general, or is it more specific to computer vision?