Neural Networks & Deep Learning
What neural networks are, why depth works, and the transformer that powers modern AI.
intermediate · 50 minutes
A practical, intermediate tour of how neural networks actually learn. We build from the single artificial neuron up through backpropagation, regularization, and the architectures (CNNs, RNNs, Transformers) that power modern AI. Expect intuition grounded in correct math, plus the training realities and misconceptions that trip up practitioners.
What you'll learn
- The Artificial Neuron and Why Nonlinearity Matters
- Layers, Depth, and the Forward Pass
- Loss Functions, Gradient Descent, and Backpropagation
- Overfitting, Regularization, and Optimization Pitfalls
- Vanishing/Exploding Gradients and Their Fixes
- A Tour of Architectures and Practical Training Realities