1 Model,contributionsandrelatedworks Randomfeaturesmodelasa2-layersneuralnetwork. Givennobservations(x1,y1), (xn,yn) withxi Rp andyi Rforeachi=1,,n,theobjectofstudyofthispaperistheestimate bα=argmin

We establish Central Limit Theorems (CLT) for the derivatives of 2-layers NN models in(2) when n,p,d + in the proportional asymptotic regime(6). A weighted average of the gradients of the trained NN, up to an explicit additive correction, is proved to be asymptotically normal, where the variance of the limit can be estimatedexplicitly.