What is a neural network?

A neural network takes in data (like pixel values of an image), passes it through layers of interconnected nodes, and produces an output (like "this is a cat"). Each connection has a "weight" that the network adjusts during training.

How training works

During training, the network makes a prediction, compares it to the correct answer, and calculates how wrong it was (the "loss"). It then adjusts its internal weights slightly to reduce that error, repeating this process over thousands or millions of examples until it gets consistently accurate.

Common types

• Feedforward networks — the simplest type, information flows one direction
• Convolutional neural networks (CNNs) — specialized for image data
• Transformers — specialized for sequential data like text, and the architecture behind most modern language models