<p>Let <i>P</i> be a set of <i>n</i> points in <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\({\mathbb {R}}^d\)</EquationSource> <EquationSource Format="MATHML"><math> <msup> <mrow> <mi mathvariant="double-struck">R</mi> </mrow> <mi>d</mi> </msup> </math></EquationSource> </InlineEquation>, in general position. We remove all of them one by one, in each step erasing one vertex of the convex hull of the current remaining set. Let <InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(g_d(P)\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <msub> <mi>g</mi> <mi>d</mi> </msub> <mrow> <mo stretchy="false">(</mo> <mi>P</mi> <mo stretchy="false">)</mo> </mrow> </mrow> </math></EquationSource> </InlineEquation> denote the number of different removal orders we can attain while erasing all points of <i>P</i> this way, and let <InlineEquation ID="IEq3"> <EquationSource Format="TEX">\(g_d(n)\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <msub> <mi>g</mi> <mi>d</mi> </msub> <mrow> <mo stretchy="false">(</mo> <mi>n</mi> <mo stretchy="false">)</mo> </mrow> </mrow> </math></EquationSource> </InlineEquation> be the <i>minimum</i> of <InlineEquation ID="IEq4"> <EquationSource Format="TEX">\(g_d(P)\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <msub> <mi>g</mi> <mi>d</mi> </msub> <mrow> <mo stretchy="false">(</mo> <mi>P</mi> <mo stretchy="false">)</mo> </mrow> </mrow> </math></EquationSource> </InlineEquation> over all <i>n</i>-element point sets <InlineEquation ID="IEq5"> <EquationSource Format="TEX">\(P\subset {\mathbb {R}}^d\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>P</mi> <mo>⊂</mo> <msup> <mrow> <mi mathvariant="double-struck">R</mi> </mrow> <mi>d</mi> </msup> </mrow> </math></EquationSource> </InlineEquation>. Dumitrescu and Tóth showed that <InlineEquation ID="IEq6"> <EquationSource Format="TEX">\(g_d(n)\le (d+1)^{(d+1)^2n}\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <msub> <mi>g</mi> <mi>d</mi> </msub> <mrow> <mo stretchy="false">(</mo> <mi>n</mi> <mo stretchy="false">)</mo> </mrow> <mo>≤</mo> <msup> <mrow> <mo stretchy="false">(</mo> <mi>d</mi> <mo>+</mo> <mn>1</mn> <mo stretchy="false">)</mo> </mrow> <mrow> <msup> <mrow> <mo stretchy="false">(</mo> <mi>d</mi> <mo>+</mo> <mn>1</mn> <mo stretchy="false">)</mo> </mrow> <mn>2</mn> </msup> <mi>n</mi> </mrow> </msup> </mrow> </math></EquationSource> </InlineEquation>. We substantially improve their bound, by proving that <InlineEquation ID="IEq7"> <EquationSource Format="TEX">\(g_d(n)=O((d+d\ln {d})^{(2+\frac{(d-1)}{\lfloor d\ln {d}\rfloor })n})\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <msub> <mi>g</mi> <mi>d</mi> </msub> <mrow> <mo stretchy="false">(</mo> <mi>n</mi> <mo stretchy="false">)</mo> </mrow> <mo>=</mo> <mi>O</mi> <mrow> <mo stretchy="false">(</mo> <msup> <mrow> <mo stretchy="false">(</mo> <mi>d</mi> <mo>+</mo> <mi>d</mi> <mo>ln</mo> <mi>d</mi> <mo stretchy="false">)</mo> </mrow> <mrow> <mo stretchy="false">(</mo> <mn>2</mn> <mo>+</mo> <mfrac> <mrow> <mo stretchy="false">(</mo> <mi>d</mi> <mo>-</mo> <mn>1</mn> <mo stretchy="false">)</mo> </mrow> <mrow> <mo>⌊</mo> <mi>d</mi> <mo>ln</mo> <mi>d</mi> <mo>⌋</mo> </mrow> </mfrac> <mo stretchy="false">)</mo> <mi>n</mi> </mrow> </msup> <mo stretchy="false">)</mo> </mrow> </mrow> </math></EquationSource> </InlineEquation>. It follows that, for any <InlineEquation ID="IEq8"> <EquationSource Format="TEX">\(\epsilon &gt;0\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>ϵ</mi> <mo>&gt;</mo> <mn>0</mn> </mrow> </math></EquationSource> </InlineEquation>, there exist sufficiently high dimensional point sets <InlineEquation ID="IEq9"> <EquationSource Format="TEX">\(P\subset {\mathbb {R}}^d\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>P</mi> <mo>⊂</mo> <msup> <mrow> <mi mathvariant="double-struck">R</mi> </mrow> <mi>d</mi> </msup> </mrow> </math></EquationSource> </InlineEquation> with <InlineEquation ID="IEq10"> <EquationSource Format="TEX">\(g_d(P)\le O(d^{(2+\epsilon )n})\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <msub> <mi>g</mi> <mi>d</mi> </msub> <mrow> <mo stretchy="false">(</mo> <mi>P</mi> <mo stretchy="false">)</mo> </mrow> <mo>≤</mo> <mi>O</mi> <mrow> <mo stretchy="false">(</mo> <msup> <mi>d</mi> <mrow> <mo stretchy="false">(</mo> <mn>2</mn> <mo>+</mo> <mi>ϵ</mi> <mo stretchy="false">)</mo> <mi>n</mi> </mrow> </msup> <mo stretchy="false">)</mo> </mrow> </mrow> </math></EquationSource> </InlineEquation>. This almost closes the gap between the upper bound and the best-known lower bound <InlineEquation ID="IEq11"> <EquationSource Format="TEX">\((d+1)^n\)</EquationSource> <EquationSource Format="MATHML"><math> <msup> <mrow> <mo stretchy="false">(</mo> <mi>d</mi> <mo>+</mo> <mn>1</mn> <mo stretchy="false">)</mo> </mrow> <mi>n</mi> </msup> </math></EquationSource> </InlineEquation> for large values of <i>d</i>.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

The Minimum Number of Peeling Sequences of a Point Set

  • Dániel G. Simon

摘要

Let P be a set of n points in \({\mathbb {R}}^d\) R d , in general position. We remove all of them one by one, in each step erasing one vertex of the convex hull of the current remaining set. Let \(g_d(P)\) g d ( P ) denote the number of different removal orders we can attain while erasing all points of P this way, and let \(g_d(n)\) g d ( n ) be the minimum of \(g_d(P)\) g d ( P ) over all n-element point sets \(P\subset {\mathbb {R}}^d\) P R d . Dumitrescu and Tóth showed that \(g_d(n)\le (d+1)^{(d+1)^2n}\) g d ( n ) ( d + 1 ) ( d + 1 ) 2 n . We substantially improve their bound, by proving that \(g_d(n)=O((d+d\ln {d})^{(2+\frac{(d-1)}{\lfloor d\ln {d}\rfloor })n})\) g d ( n ) = O ( ( d + d ln d ) ( 2 + ( d - 1 ) d ln d ) n ) . It follows that, for any \(\epsilon >0\) ϵ > 0 , there exist sufficiently high dimensional point sets \(P\subset {\mathbb {R}}^d\) P R d with \(g_d(P)\le O(d^{(2+\epsilon )n})\) g d ( P ) O ( d ( 2 + ϵ ) n ) . This almost closes the gap between the upper bound and the best-known lower bound \((d+1)^n\) ( d + 1 ) n for large values of d.