Supervised Deep Learning in High Energy Phenomenology: a Mini Review *

Murat Abdughani,Jie Ren,Lei Wu,Jin-Min Yang,Jun Zhao

Fig. 11 (Color online) The likelihood functions reconstructed using the shallow neural network, the deep neural network and our machine learning scan algorithm. The first two methods use 2000 parameter samples to learn the target likelihood function directly. The machine learning scan algorithm runs until 2,000 parameter points are sampled. The colors in the figure represent the values of the reconstructed likelihood function. This figure is taken from Ref. [12].