arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2201.07329math.STstat.TH
收藏

有界凸约束下高斯序列模型的极大极小速率

On the minimax rate of the Gaussian sequence model under bounded convex constraints

Matey Neykov

摘要:

我们确定了精确的极小极大率的高斯序列模型下有界凸约束,纯粹在给定的约束集$K$的局部几何。我们的主要结果表明,在平方$\ell_2$损失下的极小极大风险(直到常数因子)由具有\begin{align*} ε^* = \sup \bigg\{ε: \frac{ε^2}{σ^2} \leq \log M^{\operatorname{loc}}(ε)\bigg\}, \end{align*}的$ε^{*2} \wedge \operatorname{diam}(K)^2$给出,其中$\log M^{\operatorname{loc}}(ε)$表示集合$K$的局部熵,$σ^2$是噪声的方差。我们利用我们的抽象结果重新推导出已知的极小极大率的一些特殊的集$K$,如超矩形,椭圆,更一般的二次凸正交对称集。最后,我们将我们的结果扩展到无界的情况下,与已知的$σ^2$表明,在这种情况下,极大极小率是$ε^{*2}$。

英文摘要:

We determine the exact minimax rate of a Gaussian sequence model under bounded convex constraints, purely in terms of the local geometry of the given constraint set $K$. Our main result shows that the minimax risk (up to constant factors) under the squared $\ell_2$ loss is given by $ε^{*2} \wedge \operatorname{diam}(K)^2$ with \begin{align*} ε^* = \sup \bigg\{ε: \frac{ε^2}{σ^2} \leq \log M^{\operatorname{loc}}(ε)\bigg\}, \end{align*} where $\log M^{\operatorname{loc}}(ε)$ denotes the local entropy of the set $K$, and $σ^2$ is the variance of the noise. We utilize our abstract result to re-derive known minimax rates for some special sets $K$ such as hyperrectangles, ellipses, and more generally quadratically convex orthosymmetric sets. Finally, we extend our results to the unbounded case with known $σ^2$ to show that the minimax rate in that case is $ε^{*2}$.

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。

↑