Deep Learning CC - rFronteddu/general_wiki GitHub Wiki

Supervised learning is a type of machine learning that trains a model on labeled data. In supervised learning, each input is associated with an output, or target, and the neural network learns by iteratively adjusting its parameters to minimize the difference between its predictions and the actual outputs.

A single neuron learns one hyperplane. Non-convex data requires multiple hyperplanes or higher dimensions.

Ask yourself:

• Can I draw one straight line that separates the classes? If not:
• ❌ Single neuron
• ❌ Logistic regression
• ❌ Linear SVM

You need:

• MLP
• Kernel method
• Feature map

Backpropagation

Backpropagation is just the chain rule applied repeatedly from the loss back to the weights.

Dense Neural Network

A dense neural network consists of multiple stacked layers of neurons. Each neuron in a layer receives inputs from all neurons in the previous layer and sends outputs to all neurons in the following layer, in a classical feed-forward architecture. Dense neural networks are also known as fully connected neural networks or multilayer perceptrons.