OpenGrok

684 the current approximation of the affine hull.
685 The list of witnesses is used to construct an initial approximation
874 we can, in the general case, only compute an approximation
879 such that $R^+ \subseteq T$.  Of course, we want this approximation
881 $R^+$ and we want to detect the cases where the approximation is
884 For computing an approximation of the transitive closure of $R$,
886 and first compute an approximation of $R^k$ for $k \ge 1$ and then project
908 \subsection{Computing an Approximation of $R^k$}
965 and then the approximation computed in \eqref{eq:transitive:approx}
1058 We will use the following approximation $Q_i$ for $P_i'$:
1114 has integer vertices, then the approximation is exact, i.e.,
1126 not compute the most accurate affine approximation of
1147 The approximation of $k \, \Delta_i(\vec s)$ can therefore be obtained
1154 of Farkas' lemma is needed to obtain the approximation of
1169 $y - 1 + n \ge 0$, leading to the following approximation of the
1208 The approximation for $k\,\Delta$,
1217 Finally, the computed approximation for $R^+$,
1330 the following approximation for $R^+$:
1343 and we obtain the approximation
1353 Note, however, that this is not the most accurate affine approximation that
1363 The approximation $T$ for the transitive closure $R^+$ can be obtained
1364 by projecting out the parameter $k$ from the approximation $K$
1390 to checking whether the approximation $K$ of the power is exact.
1418 Note that if $R^+$ is acyclic and $T$ is not, then the approximation
1419 is clearly not exact and the approximation of the power $K$
1426 then the accuracy of the approximation may be improved by computing
1427 an approximation of each strongly connected components separately.
1471 If the approximation turns out to be inexact for any of the components,
1861 or at least an approximation of this convex hull.
2016 Presumably, the authors mean that a ``d-form'' approximation

Indexes created Tue Mar 03 05:31:39 UTC 2026