<p>We show an area law in the mutual information for the maximally-mixed state <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(\Omega \)</EquationSource> <EquationSource Format="MATHML"><math> <mi mathvariant="normal">Ω</mi> </math></EquationSource> </InlineEquation> in the ground space of general Hamiltonians, which is independent of the underlying ground space degeneracy. Our result assumes the existence of a ‘good’ approximation to the ground state projector (a good AGSP), a crucial ingredient in previous area-law proofs. Such approximations have been explicitly derived for 1D gapped local Hamiltonians and 2D frustration-free locally-gapped Hamiltonians. As a corollary, we show that in 1D gapped local Hamiltonians, for any <InlineEquation ID="IEq2"> <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> and any bipartition <InlineEquation ID="IEq3"> <EquationSource Format="TEX">\(L\cup L^c\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>L</mi> <mo>∪</mo> <msup> <mi>L</mi> <mi>c</mi> </msup> </mrow> </math></EquationSource> </InlineEquation> of the system, <Equation ID="Equ58"> <EquationSource Format="TEX">\(\begin{aligned} \textrm{I}^\epsilon _{\max } \, \! \! \left( L : L^c \right) _{\Omega } \le {\textrm{O}}\left( \log (|L|\log (d))+\log (1/\epsilon )\right) , \end{aligned}\)</EquationSource> <EquationSource Format="MATHML"><math display="block"> <mrow> <mtable> <mtr> <mtd columnalign="right"> <mrow> <msubsup> <mtext>I</mtext> <mo movablelimits="true">max</mo> <mi>ϵ</mi> </msubsup> <mspace width="0.166667em" /> <mspace width="-0.166667em" /> <mspace width="-0.166667em" /> <msub> <mfenced close=")" open="("> <mi>L</mi> <mo>:</mo> <msup> <mi>L</mi> <mi>c</mi> </msup> </mfenced> <mi mathvariant="normal">Ω</mi> </msub> <mo>≤</mo> <mtext>O</mtext> <mfenced close=")" open="("> <mo>log</mo> <mo stretchy="false">(</mo> <mo stretchy="false">|</mo> <mi>L</mi> <mo stretchy="false">|</mo> <mo>log</mo> <mo stretchy="false">(</mo> <mi>d</mi> <mo stretchy="false">)</mo> <mo stretchy="false">)</mo> <mo>+</mo> <mo>log</mo> <mo stretchy="false">(</mo> <mn>1</mn> <mo stretchy="false">/</mo> <mi>ϵ</mi> <mo stretchy="false">)</mo> </mfenced> <mo>,</mo> </mrow> </mtd> </mtr> </mtable> </mrow> </math></EquationSource> </Equation>where |<i>L</i>| represents the number of sites in <i>L</i>, <i>d</i> is the dimension of a site and <InlineEquation ID="IEq4"> <EquationSource Format="TEX">\( \textrm{I}^\epsilon _{\max } \, \! \! \left( L : L^c \right) _{\Omega }\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <msubsup> <mtext>I</mtext> <mo movablelimits="true">max</mo> <mi>ϵ</mi> </msubsup> <mspace width="0.166667em" /> <mspace width="-0.166667em" /> <mspace width="-0.166667em" /> <msub> <mfenced close=")" open="("> <mi>L</mi> <mo>:</mo> <msup> <mi>L</mi> <mi>c</mi> </msup> </mfenced> <mi mathvariant="normal">Ω</mi> </msub> </mrow> </math></EquationSource> </InlineEquation> represents the <InlineEquation ID="IEq5"> <EquationSource Format="TEX">\(\epsilon \)</EquationSource> <EquationSource Format="MATHML"><math> <mi>ϵ</mi> </math></EquationSource> </InlineEquation>-<i>smoothed maximum mutual information</i> with respect to the <InlineEquation ID="IEq6"> <EquationSource Format="TEX">\(L:L^c\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>L</mi> <mo>:</mo> <msup> <mi>L</mi> <mi>c</mi> </msup> </mrow> </math></EquationSource> </InlineEquation> partition in <InlineEquation ID="IEq7"> <EquationSource Format="TEX">\(\Omega \)</EquationSource> <EquationSource Format="MATHML"><math> <mi mathvariant="normal">Ω</mi> </math></EquationSource> </InlineEquation>. From this bound we then conclude <InlineEquation ID="IEq8"> <EquationSource Format="TEX">\( \textrm{I} \, \! \! \left( L : L^c \right) _\Omega \le {\textrm{O}}\left( \log (|L|\log (d))\right) \)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mtext>I</mtext> <mspace width="0.166667em" /> <mspace width="-0.166667em" /> <mspace width="-0.166667em" /> <msub> <mfenced close=")" open="("> <mi>L</mi> <mo>:</mo> <msup> <mi>L</mi> <mi>c</mi> </msup> </mfenced> <mi mathvariant="normal">Ω</mi> </msub> <mo>≤</mo> <mtext>O</mtext> <mfenced close=")" open="("> <mo>log</mo> <mo stretchy="false">(</mo> <mo stretchy="false">|</mo> <mi>L</mi> <mo stretchy="false">|</mo> <mo>log</mo> <mo stretchy="false">(</mo> <mi>d</mi> <mo stretchy="false">)</mo> <mo stretchy="false">)</mo> </mfenced> </mrow> </math></EquationSource> </InlineEquation> – an area law for the mutual information in 1D systems with a logarithmic correction. In addition, we show that <InlineEquation ID="IEq9"> <EquationSource Format="TEX">\(\Omega \)</EquationSource> <EquationSource Format="MATHML"><math> <mi mathvariant="normal">Ω</mi> </math></EquationSource> </InlineEquation> can be approximated in trace norm up to <InlineEquation ID="IEq10"> <EquationSource Format="TEX">\(\epsilon \)</EquationSource> <EquationSource Format="MATHML"><math> <mi>ϵ</mi> </math></EquationSource> </InlineEquation> with a state of Schmidt rank of at most <InlineEquation ID="IEq11"> <EquationSource Format="TEX">\(\textrm{poly}(|L|\log (d)/\epsilon )\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mtext>poly</mtext> <mo stretchy="false">(</mo> <mo stretchy="false">|</mo> <mi>L</mi> <mo stretchy="false">|</mo> <mo>log</mo> <mo stretchy="false">(</mo> <mi>d</mi> <mo stretchy="false">)</mo> <mo stretchy="false">/</mo> <mi>ϵ</mi> <mo stretchy="false">)</mo> </mrow> </math></EquationSource> </InlineEquation>, leading to a good MPO approximation for <InlineEquation ID="IEq12"> <EquationSource Format="TEX">\(\Omega \)</EquationSource> <EquationSource Format="MATHML"><math> <mi mathvariant="normal">Ω</mi> </math></EquationSource> </InlineEquation> with polynomial bond dimension. Similar corollaries are derived for the mutual information of 2D frustration-free and locally-gapped local Hamiltonians.</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

Area Laws and Tensor Networks for Maximally Mixed Ground States

  • Itai Arad,
  • Raz Firanko,
  • Rahul Jain

摘要

We show an area law in the mutual information for the maximally-mixed state \(\Omega \) Ω in the ground space of general Hamiltonians, which is independent of the underlying ground space degeneracy. Our result assumes the existence of a ‘good’ approximation to the ground state projector (a good AGSP), a crucial ingredient in previous area-law proofs. Such approximations have been explicitly derived for 1D gapped local Hamiltonians and 2D frustration-free locally-gapped Hamiltonians. As a corollary, we show that in 1D gapped local Hamiltonians, for any \(\epsilon >0\) ϵ > 0 and any bipartition \(L\cup L^c\) L L c of the system, \(\begin{aligned} \textrm{I}^\epsilon _{\max } \, \! \! \left( L : L^c \right) _{\Omega } \le {\textrm{O}}\left( \log (|L|\log (d))+\log (1/\epsilon )\right) , \end{aligned}\) I max ϵ L : L c Ω O log ( | L | log ( d ) ) + log ( 1 / ϵ ) , where |L| represents the number of sites in L, d is the dimension of a site and \( \textrm{I}^\epsilon _{\max } \, \! \! \left( L : L^c \right) _{\Omega }\) I max ϵ L : L c Ω represents the \(\epsilon \) ϵ -smoothed maximum mutual information with respect to the \(L:L^c\) L : L c partition in \(\Omega \) Ω . From this bound we then conclude \( \textrm{I} \, \! \! \left( L : L^c \right) _\Omega \le {\textrm{O}}\left( \log (|L|\log (d))\right) \) I L : L c Ω O log ( | L | log ( d ) ) – an area law for the mutual information in 1D systems with a logarithmic correction. In addition, we show that \(\Omega \) Ω can be approximated in trace norm up to \(\epsilon \) ϵ with a state of Schmidt rank of at most \(\textrm{poly}(|L|\log (d)/\epsilon )\) poly ( | L | log ( d ) / ϵ ) , leading to a good MPO approximation for \(\Omega \) Ω with polynomial bond dimension. Similar corollaries are derived for the mutual information of 2D frustration-free and locally-gapped local Hamiltonians.