(1) The input of the neural network is x, and the expected output is H (x) .If the input x is passed to the output as the initial result, then the target to be learned at this time F (x) = H (x) -x is also The residual module
(2) Resnet learning unit is equivalent to changing the learning goal, no longer learning a complete output H (x), but learning the difference between output and input H (x) -x ---- that is Residuals
The goal of learning is residuals
