Constant-specified and exponential concentration inequalities play an essential role in the finite-sample theory of machine learning high-dimensional statistics area. We obtain sharper constants-specified for sum independent sub-Weibull random variables, which leads to a mixture two tails: sub-Gaussian small deviations large from mean. These bounds are new improve existing with constants. In ad...