Neural networks are ML models loosely inspired by the brain — layers of interconnected nodes that learn complex patterns from data.

Structure

• Input layer: Receives features
• Hidden layers: Learn increasingly abstract representations
• Output layer: Produces the prediction

How They Learn

Backpropagation — calculate the error, then adjust weights backward through the network using gradient descent.

Why Deep Learning?

• Automatically learns features from raw data (images, text, audio)
• Scales with data — more data = better performance
• Powers all modern AI: LLMs, image generation, speech recognition

Reference:

3Blue1Brown neural networks

https://www.3blue1brown.com/topics/neural-networks