Anastasis Kratsios, Valentin Debarnot, Ivan Dokmani'c
{"title":"小型变压器计算通用公制嵌入","authors":"Anastasis Kratsios, Valentin Debarnot, Ivan Dokmani'c","doi":"10.48550/arXiv.2209.06788","DOIUrl":null,"url":null,"abstract":"We study representations of data from an arbitrary metric space $\\mathcal{X}$ in the space of univariate Gaussian mixtures with a transport metric (Delon and Desolneux 2020). We derive embedding guarantees for feature maps implemented by small neural networks called \\emph{probabilistic transformers}. Our guarantees are of memorization type: we prove that a probabilistic transformer of depth about $n\\log(n)$ and width about $n^2$ can bi-H\\\"{o}lder embed any $n$-point dataset from $\\mathcal{X}$ with low metric distortion, thus avoiding the curse of dimensionality. We further derive probabilistic bi-Lipschitz guarantees, which trade off the amount of distortion and the probability that a randomly chosen pair of points embeds with that distortion. If $\\mathcal{X}$'s geometry is sufficiently regular, we obtain stronger, bi-Lipschitz guarantees for all points in the dataset. As applications, we derive neural embedding guarantees for datasets from Riemannian manifolds, metric trees, and certain types of combinatorial graphs. When instead embedding into multivariate Gaussian mixtures, we show that probabilistic transformers can compute bi-H\\\"{o}lder embeddings with arbitrarily small distortion.","PeriodicalId":14794,"journal":{"name":"J. Mach. Learn. Res.","volume":"15 1","pages":"170:1-170:48"},"PeriodicalIF":0.0000,"publicationDate":"2022-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"6","resultStr":"{\"title\":\"Small Transformers Compute Universal Metric Embeddings\",\"authors\":\"Anastasis Kratsios, Valentin Debarnot, Ivan Dokmani'c\",\"doi\":\"10.48550/arXiv.2209.06788\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"We study representations of data from an arbitrary metric space $\\\\mathcal{X}$ in the space of univariate Gaussian mixtures with a transport metric (Delon and Desolneux 2020). We derive embedding guarantees for feature maps implemented by small neural networks called \\\\emph{probabilistic transformers}. Our guarantees are of memorization type: we prove that a probabilistic transformer of depth about $n\\\\log(n)$ and width about $n^2$ can bi-H\\\\\\\"{o}lder embed any $n$-point dataset from $\\\\mathcal{X}$ with low metric distortion, thus avoiding the curse of dimensionality. We further derive probabilistic bi-Lipschitz guarantees, which trade off the amount of distortion and the probability that a randomly chosen pair of points embeds with that distortion. If $\\\\mathcal{X}$'s geometry is sufficiently regular, we obtain stronger, bi-Lipschitz guarantees for all points in the dataset. As applications, we derive neural embedding guarantees for datasets from Riemannian manifolds, metric trees, and certain types of combinatorial graphs. When instead embedding into multivariate Gaussian mixtures, we show that probabilistic transformers can compute bi-H\\\\\\\"{o}lder embeddings with arbitrarily small distortion.\",\"PeriodicalId\":14794,\"journal\":{\"name\":\"J. Mach. Learn. Res.\",\"volume\":\"15 1\",\"pages\":\"170:1-170:48\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-09-14\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"6\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"J. Mach. Learn. Res.\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.48550/arXiv.2209.06788\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"J. Mach. Learn. Res.","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.48550/arXiv.2209.06788","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 6
摘要
我们研究了任意度量空间$\mathcal{X}$中具有输运度量的单变量高斯混合空间中的数据表示(Delon and isolneux 2020)。我们推导了由称为\emph{概率转换器的小型神经网络实现的特征映射的嵌入保证}。我们的保证是记忆型的:我们证明了深度约为$n\log(n)$和宽度约为$n^2$的概率转换器可以bi-Hölder嵌入任何来自$\mathcal{X}$的具有低度量失真的$n$点数据集,从而避免了维度的诅咒。我们进一步推导了概率双lipschitz保证,它权衡了扭曲的数量和随机选择的一对点嵌入该扭曲的概率。如果$\mathcal{X}$的几何结构足够规则,我们就可以得到数据集中所有点的更强的双lipschitz保证。作为应用,我们从黎曼流形、度量树和某些类型的组合图中导出数据集的神经嵌入保证。当嵌入到多元高斯混合时,我们表明概率变压器可以计算任意小失真的bi-Hölder嵌入。
Small Transformers Compute Universal Metric Embeddings
We study representations of data from an arbitrary metric space $\mathcal{X}$ in the space of univariate Gaussian mixtures with a transport metric (Delon and Desolneux 2020). We derive embedding guarantees for feature maps implemented by small neural networks called \emph{probabilistic transformers}. Our guarantees are of memorization type: we prove that a probabilistic transformer of depth about $n\log(n)$ and width about $n^2$ can bi-H\"{o}lder embed any $n$-point dataset from $\mathcal{X}$ with low metric distortion, thus avoiding the curse of dimensionality. We further derive probabilistic bi-Lipschitz guarantees, which trade off the amount of distortion and the probability that a randomly chosen pair of points embeds with that distortion. If $\mathcal{X}$'s geometry is sufficiently regular, we obtain stronger, bi-Lipschitz guarantees for all points in the dataset. As applications, we derive neural embedding guarantees for datasets from Riemannian manifolds, metric trees, and certain types of combinatorial graphs. When instead embedding into multivariate Gaussian mixtures, we show that probabilistic transformers can compute bi-H\"{o}lder embeddings with arbitrarily small distortion.