对于浮点算法而言,许多惯用的实数算法代数规则有时并不适用。
Many of the familiar algebraic rules for real number arithmetic do not always hold for floating-point arithmetic.
数学家对此很熟悉;如果将代数中的常见问题考虑成复数而不是实数时,那么这些问题就更容易处理,尽管复数算法较难。
Mathematicians are familiar with this; common problems in algebra are more tractable when considered over the complex numbers than over the reals, even though complex arithmetic is more difficult.
答案要看 simplex 算法背后采用的代数关系,但是有关它是如何工作的解释已经超出了本文的范围。
The answer is in the algebra behind the simplex algorithm, but explaining how it works is beyond the scope of this article.
应用推荐