We obtain error approximation bounds between expected suprema of canonical processes that are generated by random vectors with independent coordinates and expected suprema of Gaussian processes. In particular, we obtain a sharper proximity estimate for Rademacher and Gaussian complexities. Our estimates are dimension-free, and depend only on the geometric parameters and the numerical complexity of the underlying index set.

翻译：我们得到了由独立坐标随机向量生成的典范过程的期望 supremum 与高斯过程的期望 supremum 之间的误差近似界。特别地，我们获得了拉德马赫复杂度与高斯复杂度之间更精确的邻近估计。我们的估计是无维的，仅依赖于底层指标集的几何参数和数值复杂度。