IEEE Standard for Interval Arithmetic
Introduction: This introduction is not part of IEEE Std 1788-2015, IEEE Standard for Interval Arithmetic. This introduction explains some of the alternative interpretations, and sometimes competing objectives, that influenced the design of this standard. Implementers should study it for a fuller understanding of the design choices made in this standard among these interpretations and objectives. For more information on interval computations, including history, applications and software, seea e.g. [B1, B14] and the references therein, and also the interval computations web site [B5]. Mathematical context Interval computation is a collaboration between human programmer and machine infrastructure which, cor- rectly done, produces mathematically proven numerical results about continuous problems—for instance, rigorous bounds on the global minimum of a function or the solution of a differential equation. It is part of the discipline of “constructive real analysis.” In the long term, the results of such computations might become sufficiently trusted to be accepted as contributing to legal decisions. The machine infrastructure acts as a body of theorems on which the correctness of an interval algorithm relies, so it must be made as reliable as is practical. In its logical chain are many links—hardware, underlying floating-point system, etc.—over which this standard has no control. The standard aims to strengthen one specific link, by defining interval objects and operations that are theoretically well-f
Contents: