Home | History | Annotate | Download | only in doc

Lines Matching defs:Parameter

20 \begin{definition}[Parameter Domain of a Set]
22 The {\em parameter domain} of $S$ is the set
47 \begin{definition}[Parameter Domain of a Relation]
49 The {\em parameter domain} of $R$ is the set
258 parameter values, i.e., the context.
310 proposed to use a ``big parameter'', say $M$, that is taken to be
328 the original problem will involve the big parameter.
329 In the original implementation of {\tt PipLib}, the big parameter could
332 implicit conditions on the big parameter through conditions on such
335 The big parameter can then never appear in any $\vec q$ because
341 add a big parameter, perform the reformulation and interpret the result
347 where it is useful to have explicit control over the big parameter,
349 of the big parameter and we believe that the user should not be bothered
355 is checked internally and a big parameter is automatically introduced when
360 parameter in the output.  Even though
361 {\tt isl} makes the same divisibility assumption on the big parameter
445 The application of parameter compression (see below)
458 Instead, {\tt isl} introduces a new parameter, say $u$, and
481 the parameter is removed.
488 \subsubsection{Parameter Compression}\label{s:compression}
492 of the parameters.  In such cases ``parameter compression''
498 the constraint $2n = 3m$ can be replaced by a single parameter $n'$
505 Although parameter compression has been implemented in {\tt isl},
638 If the parameters may be negative, then the same big parameter trick
639 used in the main tableau is applied to the context.  This big parameter
640 is of course unrelated to the big parameter from the main tableau.
641 Note that it is not a requirement for this parameter to be ``big'',
643 In {\tt PipLib}, the extra parameter is not ``big'', but this may be because
644 the big parameter of the main tableau also appears
789 by a single newly introduced parameter that represents the minimum
792 of a new parameter.  In particular, a new parameter is introduced
802 We introduce a new parameter $t$ with
887 out the parameter $k$ from the resulting relation.
1107 that for any $\vec s$ in the parameter domain of $\Delta$,
1261 $\alpha_0$ can therefore be treated as a parameter,
1364 by projecting out the parameter $k$ from the approximation $K$