Watch neurons fire, adjust weights, and see learning happen in real-time. This interactive playground lets you build intuition for how neural networks transform data layer by layer.
Neural networks are computational models inspired by biological neurons. They learn to map inputs to outputs through layers of interconnected nodes (neurons). Each connection has a weight that gets adjusted during training to minimize prediction error.
Data flows from input → hidden layers → output. Each neuron applies weights, sums inputs, and passes through an activation function.
Non-linear transformations (ReLU, Sigmoid, Tanh) that allow networks to learn complex patterns. Without them, networks would be linear.
The learning algorithm. Calculates how much each weight contributed to the error, then adjusts weights to reduce future errors.