Home▸AI & Machine Learning▸Neural Signal Propagation (2D)

Neural Signal Propagation (2D)

A leaky integrate-and-fire spiking network: membrane potential accumulates from incoming pulses and external drive, decays over time, and fires a spike — sent along every outgoing synapse — the moment it crosses threshold.

AI & Machine Learning2DModerate60 FPS📱 Mobile-adapted⇄ 3D version

This 2D companion replaces decorative pulse animation with an actual spiking-neuron model: each of the network's input, hidden and output neurons holds a membrane potential that leaks toward zero over time, is driven upward by external input or by weighted pulses arriving from upstream neurons, and fires — resetting to zero and sending a pulse down every outgoing synapse — the instant it crosses the firing threshold, so what you see on screen is the actual signal-propagation computation, not a scripted animation.

⚙ Under the hood

Leaky integrate-and-fire (LIF) spiking network across four layers (input, two hidden, output). Membrane potential V follows dV/dt = −V/τ plus incoming weighted pulses; a neuron fires when V crosses the threshold, resets to zero, and forwards a pulse scaled by each synapse's weight to every downstream neuron.

neural networkspiking neuronssignal propagationintegrate-and-firesynaptic weights

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