Blind spots of qualitative simulators

dc.contributorGraduate Program in Computer Engineering.
dc.contributor.advisorSay, Ahmet Celal Cem.
dc.contributor.authorTaşdemir, Nuri.
dc.date.accessioned2023-03-16T10:06:39Z
dc.date.available2023-03-16T10:06:39Z
dc.date.issued2007.
dc.description.abstractQualitative simulators are important tools for analyzing the possible behaviors of a dynamical system. The technique of qualitative simulation has some theoretical limitations. For some input system models, a qualitative simulator either predicts spurious (impossible) behaviors or does not predict some possible behaviors. It has been shown that a “perfect” qualitative simulator cannot be built, because there are some spurious behaviors which cannot be proven to be spurious. This thesis questions the possibility of building a qualitative simulator which can detect and eliminate all spurious behaviors which can be proven to be so. We find the answer to be negative. For each qualitative simulator which possesses some reasonable properties, there is an efficiently constructible, provably inconsistent input which cannot be rejected by the simulator in question. In addition, it is shown that there exist spurious behaviors which require exponential time to detect.
dc.format.extent30cm.
dc.format.pagesxi, 34 leaves;
dc.identifier.otherCMPE 2007 T37
dc.identifier.urihttps://digitalarchive.library.bogazici.edu.tr/handle/123456789/12521
dc.publisherThesis (M.S.)-Bogazici University. Institute for Graduate Studies in Science and Engineering, 2007.
dc.relationIncludes appendices.
dc.relationIncludes appendices.
dc.subject.lcshOptic disc.
dc.subject.lcshReasoning.
dc.subject.lcshRecursion theory.
dc.subject.lcshGödel's theorem.
dc.titleBlind spots of qualitative simulators

Files

Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
b1490332.002374.001.PDF
Size:
174.69 KB
Format:
Adobe Portable Document Format

Collections