(Machine Learning Term): One of the fundamental parameters of a neural network that is set through iterative model training. Weights refer to the strength of a particular connection between a neuron and the neuron in the next or previous layer. Weights are a parameter derived through iterative training that identifies how outputs from a node / neuron should impact the input to a subsequent node / neuron. Weights refer to the strength of a particular connection between a neuron and the neuron in the next or previous layer, or you can think of it as the probability that the node / neuron is active.