<p>Lifting theorems are used to transfer lower bounds between Boolean function complexity measures. Given a lower bound on a complexity measure <InlineEquation ID="IEq10"> <EquationSource Format="TEX">\(A\)</EquationSource> <EquationSource Format="MATHML"><math> <mi>A</mi> </math></EquationSource> </InlineEquation> for some function <InlineEquation ID="IEq11"> <EquationSource Format="TEX">\(f\)</EquationSource> <EquationSource Format="MATHML"><math> <mi>f</mi> </math></EquationSource> </InlineEquation>, we compose <InlineEquation ID="IEq12"> <EquationSource Format="TEX">\(f\)</EquationSource> <EquationSource Format="MATHML"><math> <mi>f</mi> </math></EquationSource> </InlineEquation> with a carefully chosen <i>gadget</i> function <InlineEquation ID="IEq14"> <EquationSource Format="TEX">\(g\)</EquationSource> <EquationSource Format="MATHML"><math> <mi>g</mi> </math></EquationSource> </InlineEquation> and get essentially the same lower bound on a complexity measure <InlineEquation ID="IEq3"> <EquationSource Format="TEX">\(B\)</EquationSource> <EquationSource Format="MATHML"><math> <mi>B</mi> </math></EquationSource> </InlineEquation> for the <i>lifted</i> function <InlineEquation ID="IEq4"> <EquationSource Format="TEX">\(f \diamond g\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>f</mi> <mo>⋄</mo> <mi>g</mi> </mrow> </math></EquationSource> </InlineEquation>. Lifting theorems have applications in many different areas, such as circuit complexity, communication complexity, proof complexity, etc.</p><p>One of the main questions in the context of lifting is how to choose a suitable gadget <InlineEquation ID="IEq5"> <EquationSource Format="TEX">\(g\)</EquationSource> <EquationSource Format="MATHML"><math> <mi>g</mi> </math></EquationSource> </InlineEquation>.Generally, to get better results, i.e., to minimize the losses when transferring lower bounds, we need the gadget to be of a constant size (number of inputs). Unfortunately, in many settings we know lifting results only for gadgets of size that grows with the size of <InlineEquation ID="IEq6"> <EquationSource Format="TEX">\(f\)</EquationSource> <EquationSource Format="MATHML"><math> <mi>f</mi> </math></EquationSource> </InlineEquation>, and it is unclear whether they can be improved to constant-size gadgets. This motivates us to identify the properties of gadgets that make lifting possible.</p><p>In this paper, we systematically study the question: ‘For which gadgetsdoes the lifting result hold?’ in the following four settings: lifting from decision tree depth to decision tree size, lifting from conjunction DAG width to conjunction DAG size,lifting from decision tree depth to parity decision tree depth and size, and lifting from block sensitivity to deterministic and randomized communication complexities. In all the cases, we prove the complete classification of gadgets by exposing the properties of gadgets that make lifting results hold. The structure of the results shows that there are no intermediate cases—for every gadget, there is either a polynomial lifting or no lifting at all. As a byproduct of our studies, we prove the log-rank conjecture for the class of functions that can be represented as <InlineEquation ID="IEq7"> <EquationSource Format="TEX">\(f\diamond OR \diamond XOR\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>f</mi> <mo>⋄</mo> <mi>O</mi> <mi>R</mi> <mo>⋄</mo> <mi>X</mi> <mi>O</mi> <mi>R</mi> </mrow> </math></EquationSource> </InlineEquation> for some function <InlineEquation ID="IEq8"> <EquationSource Format="TEX">\(f\)</EquationSource> <EquationSource Format="MATHML"><math> <mi>f</mi> </math></EquationSource> </InlineEquation>.</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

Lifting Dichotomies

  • Yaroslav Alekseev ,
  • Yuval Filmus,
  • Alexander V. Smal

摘要

Lifting theorems are used to transfer lower bounds between Boolean function complexity measures. Given a lower bound on a complexity measure \(A\) A for some function \(f\) f , we compose \(f\) f with a carefully chosen gadget function \(g\) g and get essentially the same lower bound on a complexity measure \(B\) B for the lifted function \(f \diamond g\) f g . Lifting theorems have applications in many different areas, such as circuit complexity, communication complexity, proof complexity, etc.

One of the main questions in the context of lifting is how to choose a suitable gadget \(g\) g .Generally, to get better results, i.e., to minimize the losses when transferring lower bounds, we need the gadget to be of a constant size (number of inputs). Unfortunately, in many settings we know lifting results only for gadgets of size that grows with the size of \(f\) f , and it is unclear whether they can be improved to constant-size gadgets. This motivates us to identify the properties of gadgets that make lifting possible.

In this paper, we systematically study the question: ‘For which gadgetsdoes the lifting result hold?’ in the following four settings: lifting from decision tree depth to decision tree size, lifting from conjunction DAG width to conjunction DAG size,lifting from decision tree depth to parity decision tree depth and size, and lifting from block sensitivity to deterministic and randomized communication complexities. In all the cases, we prove the complete classification of gadgets by exposing the properties of gadgets that make lifting results hold. The structure of the results shows that there are no intermediate cases—for every gadget, there is either a polynomial lifting or no lifting at all. As a byproduct of our studies, we prove the log-rank conjecture for the class of functions that can be represented as \(f\diamond OR \diamond XOR\) f O R X O R for some function \(f\) f .