Home▸Networks & Graph Theory▸Machine Learning Neural Network Simulation – Weight Noise Injection

🧪 Machine Learning Neural Network Simulation – Weight Noise Injection

Interactive neural network simulation – learn how artificial neurons and reinforcement learning work.

Networks & Graph Theory2DModerate60 FPS

⚙ Under the hood

This simulation allows you to visually explore the inner workings of a basic artificial neural network. By adjusting parameters like weights and activation functions, you can observe how the network learns patterns from simulated data, demonstrating core concepts of supervised learning. A weight-noise control adds random perturbations to weight updates during backpropagation. Adds a weight-noise control that perturbs weight updates during backpropagation.

Supervised LearningUnsupervised

2D · HTML5 Canvas 2D · 60 FPS target · runs fully client-side, no install