Deep Learning
How neural networks actually learn, from perceptrons and backpropagation to the architectures behind modern deep learning.
Foundations of Deep Learning
Post 1•Ch.1: What is Deep Learning? Deep Learning vs Machine Learning
Where deep learning sits inside AI and machine learning, what representation learning means and why layers learn edges before faces, five practical differences from classical machine learning (data, hardware, training time, features, interpretability), and the five forces behind deep learning's rise after 2010.
- Post 2•
Ch.2: Types of Neural Networks, History and Applications of Deep Learning
The five neural network families you'll meet first (MLP, CNN, RNN, autoencoders, GANs) and what each is built for, the boom-and-bust history from Rosenblatt's perceptron through two AI winters to the 2012 ImageNet breakthrough, and where deep learning shows up today.
Perceptron
Post 3•Ch.3: What is a Perceptron? Perceptron vs Neuron and Geometric Intuition
The perceptron, the building block of every neural network: inputs, weights, bias, a weighted sum and a step activation. How it predicts, how it compares to a biological neuron, why it is really just a line (or plane, or hyperplane) splitting space into two regions, and what that means when you run it in scikit-learn.