Carlin Wiegner > AI > Learn

• Software 1.0 (telling) versus Software 2.0 (showing) - Why Neural Nets may replace traditional programming?
• Often for complex problems you know the result you want but how of the best way to break down the problem into little steps or even how those little steps should work.

• Why neural networks can learn (almost) anything? - Why do neural networks work?
• Universal approximation thereom

• Deep Learning Chapter 1: But what is a neural network?
• Deep Learning Chapter 2: Gradient descent, how neural networks learn
• Deep Learning Chapter 3: What is backpropagation really doing?
• Deep Learning Chapter 5: But what is a GPT? Visual intro to transformers
• Deep Learning Chapter 6: Visualizing Attention, a Transformer's Heart

Created: February 29 2024.
Modified: October 24 2024.