ملخص
We present a near linear time algorithm for constructing hierarchical nets in finite metric spaces with constant doubling dimension. This data-structure is then applied to obtain improved algorithms for the following problems: Approximate nearest neighbor search, well-separated pair decomposition, spanner construction, compact representation scheme, doubling measure, and computation of the (approximate) Lipschitz constant of a function. In all cases, the running (preprocessing) time is near-linear and the space being used is linear.
اللغة الأصلية | الإنجليزيّة |
---|---|
الصفحات | 150-158 |
عدد الصفحات | 9 |
المعرِّفات الرقمية للأشياء | |
حالة النشر | نُشِر - 2005 |
منشور خارجيًا | نعم |
الحدث | 21st Annual Symposium on Computational Geometry, SCG'05 - Pisa, إيطاليا المدة: ٦ يونيو ٢٠٠٥ → ٨ يونيو ٢٠٠٥ |
!!Conference
!!Conference | 21st Annual Symposium on Computational Geometry, SCG'05 |
---|---|
الدولة/الإقليم | إيطاليا |
المدينة | Pisa |
المدة | ٦/٠٦/٠٥ → ٨/٠٦/٠٥ |