ONLY ONE OF THE OPTIONS is correct.

Consider a node in a cemputational graph in which the input is a one-dimersional Oausion distribution function, represerted by (,) where is mean and is standart deviation. The output of this node is a scalar z sampled from this distribution. Find the local gradient (z,z) Choice 1: (0,1) Choice 2: (,1) Choice 3: (,) Choice 4: (1,E) Choice 5: (1,0) Choice 6: (1,1) Choice 7: (1,t2) Choice 8: (2,1) Remark: E stands for the random veriable sampled frem wnit Goussian to be used for "reparameterization trick