我们研究在对抗性腐败下学习通用线性模型的问题。我们分析了一种经典的启发式,称为迭代修剪的最大似然估计量,该估计量已知在实践中有效地抵抗标签腐败。在标签腐败下,我们证明了这个简单的估计器在各种广泛的线性模型上实现了最小风险,包括高斯回归,泊松回归和二项式回归。最后,我们将估计器扩展到标签和协变量腐败的更具挑战性的设置,并在该环境中证明其稳健性和最佳性。
translated by 谷歌翻译
This paper presents SVAM (Sequential Variance-Altered MLE), a unified framework for learning generalized linear models under adversarial label corruption in training data. SVAM extends to tasks such as least squares regression, logistic regression, and gamma regression, whereas many existing works on learning with label corruptions focus only on least squares regression. SVAM is based on a novel variance reduction technique that may be of independent interest and works by iteratively solving weighted MLEs over variance-altered versions of the GLM objective. SVAM offers provable model recovery guarantees superior to the state-of-the-art for robust regression even when a constant fraction of training labels are adversarially corrupted. SVAM also empirically outperforms several existing problem-specific techniques for robust regression and classification. Code for SVAM is available at https://github.com/purushottamkar/svam/
translated by 谷歌翻译
我们介绍了一个普遍的框架,用于表征差异隐私保证的统计估算问题的统计效率。我们的框架,我们呼叫高维建议 - 试验释放(HPTR),在三个重要组件上建立:指数机制,强大的统计和提议 - 试验释放机制。将所有这些粘在一起是恢复力的概念,这是强大的统计估计的核心。弹性指导算法的设计,灵敏度分析和试验步骤的成功概率分析。关键识别是,如果我们设计了一种仅通过一维鲁棒统计数据访问数据的指数机制,则可以大大减少所产生的本地灵敏度。使用弹性,我们可以提供紧密的本地敏感界限。这些紧张界限在几个案例中容易转化为近乎最佳的实用程序。我们给出了将HPTR应用于统计估计问题的给定实例的一般配方,并在平均估计,线性回归,协方差估计和主成分分析的规范问题上证明了它。我们介绍了一般的公用事业分析技术,证明了HPTR几乎在文献中研究的若干场景下实现了最佳的样本复杂性。
translated by 谷歌翻译
本文研究了具有对抗性误差的强大一位压缩感应的二进制分类。假设该模型过度分配,并且感兴趣的参数有效稀疏。adaboost被考虑,并且通过其与MAX - $ \ ell_1 $ -Margin-Scressifir的关系,派生预测错误界限。开发的理论是一般的,并且允许重型的特征分布,只需要一个薄弱的时刻假设和抗浓缩条件。当特征满足小偏差下限时,示出了改善的收敛速率。特别是,结果提供了解释为什么内插对抗性噪声对于分类问题可以是无害的。模拟说明了所提出的理论。
translated by 谷歌翻译
在共享数据的统计学习和分析中,在联合学习和元学习等平台上越来越广泛地采用,有两个主要问题:隐私和鲁棒性。每个参与的个人都应该能够贡献,而不会担心泄露一个人的敏感信息。与此同时,系统应该在恶意参与者的存在中插入损坏的数据。最近的算法在学习中,学习共享数据专注于这些威胁中的一个,使系统容易受到另一个威胁。我们弥合了这个差距,以获得估计意思的规范问题。样品。我们介绍了素数,这是第一算法,实现了各种分布的隐私和鲁棒性。我们通过新颖的指数时间算法进一步补充了这一结果,提高了素数的样本复杂性,实现了近最优保证并匹配(非鲁棒)私有平均估计的已知下限。这证明没有额外的统计成本同时保证隐私和稳健性。
translated by 谷歌翻译
Popular iterative algorithms such as boosting methods and coordinate descent on linear models converge to the maximum $\ell_1$-margin classifier, a.k.a. sparse hard-margin SVM, in high dimensional regimes where the data is linearly separable. Previous works consistently show that many estimators relying on the $\ell_1$-norm achieve improved statistical rates for hard sparse ground truths. We show that surprisingly, this adaptivity does not apply to the maximum $\ell_1$-margin classifier for a standard discriminative setting. In particular, for the noiseless setting, we prove tight upper and lower bounds for the prediction error that match existing rates of order $\frac{\|\wgt\|_1^{2/3}}{n^{1/3}}$ for general ground truths. To complete the picture, we show that when interpolating noisy observations, the error vanishes at a rate of order $\frac{1}{\sqrt{\log(d/n)}}$. We are therefore first to show benign overfitting for the maximum $\ell_1$-margin classifier.
translated by 谷歌翻译
We study the fundamental task of outlier-robust mean estimation for heavy-tailed distributions in the presence of sparsity. Specifically, given a small number of corrupted samples from a high-dimensional heavy-tailed distribution whose mean $\mu$ is guaranteed to be sparse, the goal is to efficiently compute a hypothesis that accurately approximates $\mu$ with high probability. Prior work had obtained efficient algorithms for robust sparse mean estimation of light-tailed distributions. In this work, we give the first sample-efficient and polynomial-time robust sparse mean estimator for heavy-tailed distributions under mild moment assumptions. Our algorithm achieves the optimal asymptotic error using a number of samples scaling logarithmically with the ambient dimension. Importantly, the sample complexity of our method is optimal as a function of the failure probability $\tau$, having an additive $\log(1/\tau)$ dependence. Our algorithm leverages the stability-based approach from the algorithmic robust statistics literature, with crucial (and necessary) adaptations required in our setting. Our analysis may be of independent interest, involving the delicate design of a (non-spectral) decomposition for positive semi-definite matrices satisfying certain sparsity properties.
translated by 谷歌翻译
我们开发机器以设计有效的可计算和一致的估计,随着观察人数而达到零的估计误差,因为观察的次数增长,当面对可能损坏的答复,除了样本的所有品,除了每种量之外的ALL。作为具体示例,我们调查了两个问题:稀疏回归和主成分分析(PCA)。对于稀疏回归,我们实现了最佳样本大小的一致性$ n \ gtrsim(k \ log d)/ \ alpha ^ $和最佳错误率$ o(\ sqrt {(k \ log d)/(n \ cdot \ alpha ^ 2))$ N $是观察人数,$ D $是尺寸的数量,$ k $是参数矢量的稀疏性,允许在数量的数量中为逆多项式进行逆多项式样品。在此工作之前,已知估计是一致的,当Inliers $ \ Alpha $ IS $ O(1 / \ log \ log n)$,即使是(非球面)高斯设计矩阵时也是一致的。结果在弱设计假设下持有,并且在这种一般噪声存在下仅被D'Orsi等人最近以密集的设置(即一般线性回归)显示。 [DNS21]。在PCA的上下文中,我们在参数矩阵上的广泛尖端假设下获得最佳错误保证(通常用于矩阵完成)。以前的作品可以仅在假设下获得非琐碎的保证,即与最基于的测量噪声以$ n $(例如,具有方差1 / n ^ 2 $的高斯高斯)。为了设计我们的估算,我们用非平滑的普通方(如$ \ ell_1 $ norm或核规范)装备Huber丢失,并以一种新的方法来分析损失的新方法[DNS21]的方法[DNS21]。功能。我们的机器似乎很容易适用于各种估计问题。
translated by 谷歌翻译
在本文中,我们提出了一种均匀抖动的一位量化方案,以进行高维统计估计。该方案包含截断,抖动和量化,作为典型步骤。作为规范示例,量化方案应用于三个估计问题:稀疏协方差矩阵估计,稀疏线性回归和矩阵完成。我们研究了高斯和重尾政权,假定重尾数据的基本分布具有有限的第二或第四刻。对于每个模型,我们根据一位量化的数据提出新的估计器。在高斯次级政权中,我们的估计器达到了对数因素的最佳最小速率,这表明我们的量化方案几乎没有额外的成本。在重尾状态下,虽然我们的估计量基本上变慢,但这些结果是在这种单位量化和重型尾部设置中的第一个结果,或者比现有可比结果表现出显着改善。此外,我们为一位压缩传感和一位矩阵完成的问题做出了巨大贡献。具体而言,我们通过凸面编程将一位压缩感传感扩展到次高斯甚至是重尾传感向量。对于一位矩阵完成,我们的方法与标准似然方法基本不同,并且可以处理具有未知分布的预量化随机噪声。提出了有关合成数据的实验结果,以支持我们的理论分析。
translated by 谷歌翻译
我们在高维批处理设置中提出了统计上健壮和计算高效的线性学习方法,其中功能$ d $的数量可能超过样本量$ n $。在通用学习环境中,我们采用两种算法,具体取决于所考虑的损失函数是否为梯度lipschitz。然后,我们将我们的框架实例化,包括几种应用程序,包括香草稀疏,群 - 帕克斯和低升级矩阵恢复。对于每种应用,这导致了有效而强大的学习算法,这些算法在重尾分布和异常值的存在下达到了近乎最佳的估计率。对于香草$ S $ -SPARSITY,我们能够以重型尾巴和$ \ eta $ - 腐败的计算成本与非企业类似物相当的计算成本达到$ s \ log(d)/n $速率。我们通过开放源代码$ \ mathtt {python} $库提供了有效的算法实现文献中提出的最新方法。
translated by 谷歌翻译
截断的线性回归是统计学中的一个经典挑战,其中$ y = w^t x + \ varepsilon $及其相应的功能向量,$ x \ in \ mathbb {r}^k $,仅在当时才观察到标签属于某些子集$ s \ subseteq \ mathbb {r} $;否则,对$(x,y)$的存在被隐藏在观察中。以截断的观察结果的线性回归一直是其一般形式的挑战,因为〜\ citet {tobin1958估计,amemiya1973 reflecression}的早期作品。当误差的分布与已知方差正常时,〜\ citet {daskalakis2019 truncatedRegerse}的最新工作在线性模型$ w $上提供了计算和统计上有效的估计器。在本文中,当噪声方差未知时,我们为截断的线性回归提供了第一个计算和统计上有效的估计器,同时估计了噪声的线性模型和方差。我们的估计器基于对截短样品的负模样中的预测随机梯度下降的有效实施。重要的是,我们表明我们的估计错误是渐近正常的,我们使用它来为我们的估计提供明确的置信区域。
translated by 谷歌翻译
We consider the problem of robustly testing the norm of a high-dimensional sparse signal vector under two different observation models. In the first model, we are given $n$ i.i.d. samples from the distribution $\mathcal{N}\left(\theta,I_d\right)$ (with unknown $\theta$), of which a small fraction has been arbitrarily corrupted. Under the promise that $\|\theta\|_0\le s$, we want to correctly distinguish whether $\|\theta\|_2=0$ or $\|\theta\|_2>\gamma$, for some input parameter $\gamma>0$. We show that any algorithm for this task requires $n=\Omega\left(s\log\frac{ed}{s}\right)$ samples, which is tight up to logarithmic factors. We also extend our results to other common notions of sparsity, namely, $\|\theta\|_q\le s$ for any $0 < q < 2$. In the second observation model that we consider, the data is generated according to a sparse linear regression model, where the covariates are i.i.d. Gaussian and the regression coefficient (signal) is known to be $s$-sparse. Here too we assume that an $\epsilon$-fraction of the data is arbitrarily corrupted. We show that any algorithm that reliably tests the norm of the regression coefficient requires at least $n=\Omega\left(\min(s\log d,{1}/{\gamma^4})\right)$ samples. Our results show that the complexity of testing in these two settings significantly increases under robustness constraints. This is in line with the recent observations made in robust mean testing and robust covariance testing.
translated by 谷歌翻译
我们研究了在存在$ \ epsilon $ - 对抗异常值的高维稀疏平均值估计的问题。先前的工作为此任务获得了该任务的样本和计算有效算法,用于辅助性Subgaussian分布。在这项工作中,我们开发了第一个有效的算法,用于强大的稀疏平均值估计,而没有对协方差的先验知识。对于$ \ Mathbb r^d $上的分布,带有“认证有限”的$ t $ tum-矩和足够轻的尾巴,我们的算法达到了$ o(\ epsilon^{1-1/t})$带有样品复杂性$的错误(\ epsilon^{1-1/t}) m =(k \ log(d))^{o(t)}/\ epsilon^{2-2/t} $。对于高斯分布的特殊情况,我们的算法达到了$ \ tilde o(\ epsilon)$的接近最佳错误,带有样品复杂性$ m = o(k^4 \ mathrm {polylog}(d)(d))/\ epsilon^^ 2 $。我们的算法遵循基于方形的总和,对算法方法的证明。我们通过统计查询和低度多项式测试的下限来补充上限,提供了证据,表明我们算法实现的样本时间 - 错误权衡在质量上是最好的。
translated by 谷歌翻译
This paper studies the quantization of heavy-tailed data in some fundamental statistical estimation problems, where the underlying distributions have bounded moments of some order. We propose to truncate and properly dither the data prior to a uniform quantization. Our major standpoint is that (near) minimax rates of estimation error are achievable merely from the quantized data produced by the proposed scheme. In particular, concrete results are worked out for covariance estimation, compressed sensing, and matrix completion, all agreeing that the quantization only slightly worsens the multiplicative factor. Besides, we study compressed sensing where both covariate (i.e., sensing vector) and response are quantized. Under covariate quantization, although our recovery program is non-convex because the covariance matrix estimator lacks positive semi-definiteness, all local minimizers are proved to enjoy near optimal error bound. Moreover, by the concentration inequality of product process and covering argument, we establish near minimax uniform recovery guarantee for quantized compressed sensing with heavy-tailed noise.
translated by 谷歌翻译
我们在对数损失下引入条件密度估计的过程,我们调用SMP(样本Minmax预测器)。该估算器最大限度地减少了统计学习的新一般过度风险。在标准示例中,此绑定量表为$ d / n $,$ d $ d $模型维度和$ n $ sample大小,并在模型拼写条目下批判性仍然有效。作为一个不当(超出型号)的程序,SMP在模型内估算器(如最大似然估计)的内部估算器上,其风险过高的风险降低。相比,与顺序问题的方法相比,我们的界限删除了SubOltimal $ \ log n $因子,可以处理无限的类。对于高斯线性模型,SMP的预测和风险受到协变量的杠杆分数,几乎匹配了在没有条件的线性模型的噪声方差或近似误差的条件下匹配的最佳风险。对于Logistic回归,SMP提供了一种非贝叶斯方法来校准依赖于虚拟样本的概率预测,并且可以通过解决两个逻辑回归来计算。它达到了$ O的非渐近风险((d + b ^ 2r ^ 2)/ n)$,其中$ r $绑定了特征的规范和比较参数的$ B $。相比之下,在模型内估计器内没有比$ \ min达到更好的速率({b r} / {\ sqrt {n}},{d e ^ {br} / {n})$。这为贝叶斯方法提供了更实用的替代方法,这需要近似的后部采样,从而部分地解决了Foster等人提出的问题。 (2018)。
translated by 谷歌翻译
In large-scale distributed learning, security issues have become increasingly important. Particularly in a decentralized environment, some computing units may behave abnormally, or even exhibit Byzantine failures-arbitrary and potentially adversarial behavior. In this paper, we develop distributed learning algorithms that are provably robust against such failures, with a focus on achieving optimal statistical performance. A main result of this work is a sharp analysis of two robust distributed gradient descent algorithms based on median and trimmed mean operations, respectively. We prove statistical error rates for three kinds of population loss functions: strongly convex, nonstrongly convex, and smooth non-convex. In particular, these algorithms are shown to achieve order-optimal statistical error rates for strongly convex losses. To achieve better communication efficiency, we further propose a median-based distributed algorithm that is provably robust, and uses only one communication round. For strongly convex quadratic loss, we show that this algorithm achieves the same optimal error rate as the robust distributed gradient descent algorithms.
translated by 谷歌翻译
We establish a simple connection between robust and differentially-private algorithms: private mechanisms which perform well with very high probability are automatically robust in the sense that they retain accuracy even if a constant fraction of the samples they receive are adversarially corrupted. Since optimal mechanisms typically achieve these high success probabilities, our results imply that optimal private mechanisms for many basic statistics problems are robust. We investigate the consequences of this observation for both algorithms and computational complexity across different statistical problems. Assuming the Brennan-Bresler secret-leakage planted clique conjecture, we demonstrate a fundamental tradeoff between computational efficiency, privacy leakage, and success probability for sparse mean estimation. Private algorithms which match this tradeoff are not yet known -- we achieve that (up to polylogarithmic factors) in a polynomially-large range of parameters via the Sum-of-Squares method. To establish an information-computation gap for private sparse mean estimation, we also design new (exponential-time) mechanisms using fewer samples than efficient algorithms must use. Finally, we give evidence for privacy-induced information-computation gaps for several other statistics and learning problems, including PAC learning parity functions and estimation of the mean of a multivariate Gaussian.
translated by 谷歌翻译
元学习或学习学习,寻求设计算法,可以利用以前的经验快速学习新技能或适应新环境。表示学习 - 用于执行元学习的关键工具 - 了解可以在多个任务中传输知识的数据表示,这在数据稀缺的状态方面是必不可少的。尽管最近在Meta-Leature的实践中感兴趣的兴趣,但缺乏元学习算法的理论基础,特别是在学习可转让陈述的背景下。在本文中,我们专注于多任务线性回归的问题 - 其中多个线性回归模型共享常见的低维线性表示。在这里,我们提供了可提供的快速,采样高效的算法,解决了(1)的双重挑战,从多个相关任务和(2)将此知识转移到新的,看不见的任务中的常见功能。两者都是元学习的一般问题的核心。最后,我们通过在学习这些线性特征的样本复杂性上提供信息定理下限来补充这些结果。
translated by 谷歌翻译
在本文中,我们研究了经验$ \ ell_2 $最小化(erm)的估计性能(标准)阶段检索(NPR),由$ y_k = | \ alpha_k^*x_0 |^2+\ eta_k $,或嘈杂的广义阶段检索(NGPR)以$ y_k = x_0^*a_kx_0 + \ eta_k $,其中$ x_0 \ in \ mathbb {k}^d $是所需的信号,$ n $是样本大小,$ \ eta =(\ eta_1,...,\ eta_n)^\ top $是噪声向量。我们在不同的噪声模式下建立了新的错误界限,我们的证明对$ \ mathbb {k} = \ mathbb {r} $和$ \ mathbb {k} = \ mathbb {c} $有效。在任意噪声向量$ \ eta $下的NPR中,我们得出了一个新的错误$ o \ big(\ | \ eta \ | _ \ | _ \ infty \ sqrt {\ frac {d} {1}^\ top \ eta |} {n} \ big)$,它比当前已知的一个$ o \ big(\ frac {\ | \ eTa \ |} {\ sqrt {\ sqrt {n}} \ big big )$在许多情况下。在NGPR中,我们显示了$ o \ big(\ | \ eta \ | \ frac {\ sqrt {d}}} {n} {n} \ big)$ for nutary $ \ eta $。在这两个问题上,任意噪声的范围立即引起$ \ tilde {o}(\ sqrt {\ frac {d} {n}}}})$,用于次高斯或次指数随机噪声,带有一些常规但不可吻的去除或削弱的假设(例如,独立或均值均值的条件)。此外,我们首次尝试在假定$ l $ -th时刻的重尾随机噪声下进行ERM。为了实现偏见和差异之间的权衡,我们截断了响应并提出了相应的稳健ERM估计器,该估计量具有保证$ \ tilde {o} \ big(\ big [\ sqrt {\ frac {\ frac {d}) {n}} \ big]^{1-1/l} \ big)$在NPR,NGPR中。所有错误都直接扩展到等级$ r $矩阵恢复的更普遍的问题,这些结果得出的结论是,全级框架$ \ {a_k \} _ {k = 1}^n $ in ngpr是比级别1帧$ \ {\ alpha_k \ alpha_k^*\} _ {k = 1}^n $在npr中更强大。提出了广泛的实验结果,以说明我们的理论发现。
translated by 谷歌翻译
在本文中,我们研究了代理人(个人)具有战略性或自我利益的情况,并且在报告数据时关注其隐私。与经典环境相比,我们的目标是设计机制,这些机制既可以激励大多数代理来真实地报告他们的数据并保留个人报告的隐私,而它们的输出也应接近基础参数。在本文的第一部分中,我们考虑了协变量是次高斯的情况,并且在他们只有有限的第四瞬间的情况下进行了重尾。首先,我们是受可能性功能最大化器的固定条件的动机,我们得出了一种新颖的私人和封闭式估计量。基于估算器,我们提出了一种机制,该机制通过对几种规范模型的计算和付款方案进行一些适当的设计具有以下属性,例如线性回归,逻辑回归和泊松回归:(1)机制为$ O(1) $ - 接点差异私有(概率至少$ 1-O(1)$); (2)这是一个$ o(\ frac {1} {n})$ - 近似于$(1-o(1))$的代理的近似贝叶斯nash平衡,以真实地报告其数据,其中$ n $是代理人的数量; (3)输出可能会达到基础参数的$ O(1)$; (4)对于机制中的$(1-o(1))$的代理分数是个人合理的; (5)分析师运行该机制所需的付款预算为$ O(1)$。在第二部分中,我们考虑了在更通用的环境下的线性回归模型,在该设置中,协变量和响应都是重尾,只有有限的第四次矩。通过使用$ \ ell_4 $ -norm收缩运算符,我们提出了一种私人估算器和付款方案,该方案具有与次高斯案例相似的属性。
translated by 谷歌翻译