# Deep Learning Notes Hub > [!Note] > Hub consolidating notes on **Deep Learning** — from perceptron foundations > to modern architectures. Reference: Goodfellow, Bengio & Courville (*Deep > Learning*). ## 1. Neuron-Level Foundations - [[Neuron]] - [[Perceptron]] - [[Sigmoid Neuron]] - [[RELU Neuron]] - [[Activation function]] - [[Input or combination function]] - [[Perceptron training]] - [[Gradient descent method for Perceptron]] --- ## 2. Network Architecture - [[Neural Network Architecture]] - [[Multilayer Perceptron]] - [[Dimensions of a neural network]] - [[Dimension and topology hidden layer]] - [[Universal Approximation Theorem]] --- ## 3. Training - [[Backpropagation]] - [[Vanishing and Exploding Gradients]] - [[Weight Initialization]] - [[Optimizers SGD Adam]] - [[Learning Rate Schedules]] - [[Cross-Entropy Loss]] - [[Early Stopping]] --- ## 4. Regularization - [[Dropout]] - [[Batch Normalization]] - [[Layer Normalization]] - [[Skip Connections]] --- ## 5. Convolutional and Recurrent Architectures - [[Convolutional Neural Network]] - [[Convolution and Pooling]] - [[Recurrent Neural Network]] - [[Long Short-Term Memory]] - [[Gated Recurrent Unit]] - [[Graph Neural Network]] *(Transformer is in [[5 - LLM Foundations Hub]].)* --- ## 6. Generative Models - [[Autoencoder]] - [[Variational Autoencoder]] - [[Generative Adversarial Network]] *(Diffusion Model is in [[5 - LLM Foundations Hub]].)*