Was ist Forward Propagation

Definition

Forward Propagation is the process within a neural network where input data is passed forward, layer by layer, through the network to ultimately generate a prediction or output.

Explanation

During this process, data flows from the input layer, through the hidden layers, and finally to the output layer. Within each neuron, input values are multiplied by their respective weights, summed up, and then passed through an activation function. Upon completion of forward propagation, the difference between the calculated result and the actual target value is determined to ascertain the model's error (or 'loss'). In real-world B2B SaaS scenarios, this is the stage where the trained model processes data in operational use (inference) — for instance, when the system evaluates a newly captured lead and calculates a real-time purchase probability.