arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1609.02094cs.ITcs.CGcs.DSmath.FAmath.IT
收藏

Johnson-Lindenstrauss引理的最优性

Optimality of the Johnson-Lindenstrauss Lemma

Kasper Green Larsen, Jelani Nelson

摘要:

对于任何整数$d, n \geq 2$和$1/({\min\{n,d\}})^{0.4999} < \varepsilon<1$,我们证明了存在一组$n$向量$X\subset \mathbb{R}^d$,使得任何满足$$ \forall x,y\in X,\ (1-\varepsilon)\|x-y\|_2^2\le \|f(x)-f(y)\|_2^2 \le (1+\varepsilon)\|x-y\|_2^2 $$的嵌入$f:X\rightarrow \mathbb{R}^m$必须具有$$ m = Ω(\varepsilon^{-2} \lg n). $$。此外,我们的下限几乎适用于所有感兴趣的$\varepsilon$,因为总是有一个等距嵌入到维度$\min\{d, n\}$(单位映射,或投影到$\mathop{span}(X)$)。以前,已知这样的下限仅适用于线性映射$f$,而不适用于如此宽的参数范围$\varepsilon, n, d$ [LN 16]。一般$f$的最佳已知下限是$m = Ω(\varepsilon^{-2}\lg n/\lg(1/\varepsilon))$ [Wel 74,Lev 83,Alo 03],这对于任何$\varepsilon = o(1)$都是次优的。

英文摘要:

For any integers $d, n \geq 2$ and $1/({\min\{n,d\}})^{0.4999} < \varepsilon<1$, we show the existence of a set of $n$ vectors $X\subset \mathbb{R}^d$ such that any embedding $f:X\rightarrow \mathbb{R}^m$ satisfying $$ \forall x,y\in X,\ (1-\varepsilon)\|x-y\|_2^2\le \|f(x)-f(y)\|_2^2 \le (1+\varepsilon)\|x-y\|_2^2 $$ must have $$ m = Ω(\varepsilon^{-2} \lg n). $$ This lower bound matches the upper bound given by the Johnson-Lindenstrauss lemma [JL84]. Furthermore, our lower bound holds for nearly the full range of $\varepsilon$ of interest, since there is always an isometric embedding into dimension $\min\{d, n\}$ (either the identity map, or projection onto $\mathop{span}(X)$). Previously such a lower bound was only known to hold against linear maps $f$, and not for such a wide range of parameters $\varepsilon, n, d$ [LN16]. The best previously known lower bound for general $f$ was $m = Ω(\varepsilon^{-2}\lg n/\lg(1/\varepsilon))$ [Wel74, Lev83, Alo03], which is suboptimal for any $\varepsilon = o(1)$.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑