间隔分析(或间隔结合传播,IBP)是一种流行的技术,用于验证和培训可提供稳健的深度神经网络,在可靠的机器学习领域是一个根本的挑战。然而,尽管努力实质性努力,解决了这一关键挑战的进展已经停滞不前,呼吁课程是间隔算术是否是前进的可行路径。本文在分析神经网络的间隔算法的局限上,我们提出了两个基本结果。我们的主要不可能性定理表明,对于任何神经网络分类只是三个点,在这些点上有一个有效的规格,间隔分析无法证明。此外,在一个隐藏层神经网络的禁用情况下,我们显示出更强烈的不可能性结果:给定任何RADIUS $ \ alpha <1 $,有一组$ O(\ alpha ^ { - 1})$积分强大的RADIUS $ \ Alpha $,以2美元分开,可以证明没有一个隐式层网络通过间隔分析稳健地进行分类。
translated by 谷歌翻译
我们提出了一个域理论框架,用于验证神经网络的鲁棒性分析。我们首先分析一般网络类别的全球鲁棒性。然后,利用Edalat的域理论L衍生物与Clarke的广义梯度相吻合的事实,我们扩展了攻击性不足的局部鲁棒性分析的框架。我们的框架是设计构造正确的算法的理想选择。我们通过开发经过验证的算法来估计前馈回归器常数来体现这一主张。我们证明了算法在可区分网络上以及一般位置relu网络的完整性。我们在有效给定域的框架内获得可计算结果。使用我们的域模型,可以统一分析可区分和非差异网络。我们使用任意推测间隔算术实施算法,并介绍一些实验的结果。我们的实现也得到了真正的验证,因为它也处理浮点错误。
translated by 谷歌翻译
We present AI 2 , the first sound and scalable analyzer for deep neural networks. Based on overapproximation, AI 2 can automatically prove safety properties (e.g., robustness) of realistic neural networks (e.g., convolutional neural networks).The key insight behind AI 2 is to phrase reasoning about safety and robustness of neural networks in terms of classic abstract interpretation, enabling us to leverage decades of advances in that area. Concretely, we introduce abstract transformers that capture the behavior of fully connected and convolutional neural network layers with rectified linear unit activations (ReLU), as well as max pooling layers. This allows us to handle real-world neural networks, which are often built out of those types of layers.We present a complete implementation of AI 2 together with an extensive evaluation on 20 neural networks. Our results demonstrate that: (i) AI 2 is precise enough to prove useful specifications (e.g., robustness), (ii) AI 2 can be used to certify the effectiveness of state-of-the-art defenses for neural networks, (iii) AI 2 is significantly faster than existing analyzers based on symbolic analysis, which often take hours to verify simple fully connected networks, and (iv) AI 2 can handle deep convolutional networks, which are beyond the reach of existing methods.
translated by 谷歌翻译
本文开发了简单的前馈神经网络,实现了所有连续功能的通用近似性,具有固定的有限数量的神经元。这些神经网络很简单,因为它们的设计具有简单且可增加的连续激活功能$ \ Sigma $利用三角波函数和软片功能。我们证明了$ \ Sigma $ -Activated网络,宽度为36d $ 36d(2d + 1)$和11 $ 11 $可以在任意小错误中估计$ d $ -dimensioanl超级函数上的任何连续功能。因此,对于监督学习及其相关的回归问题,这些网络产生的假设空间,尺寸不小于36d(2d + 1)\ times 11 $的持续功能的空间。此外,由图像和信号分类引起的分类函数在$ \ sigma $ -activated网络生成的假设空间中,宽度为36d(2d + 1)$和12 $ 12 $,当存在$ \的成对不相交的界限子集时mathbb {r} ^ d $,使得同一类的样本位于同一子集中。
translated by 谷歌翻译
深度神经网络的鲁棒性对于现代AI支持系统至关重要,应正式验证。在广泛的应用中采用了类似乙状结肠的神经网络。由于它们的非线性,通常会过度评估乙状结肠样激活功能,以进行有效的验证,这不可避免地引入了不精确度。已大量的努力致力于找到所谓的更紧密的近似值,以获得更精确的验证结果。但是,现有的紧密定义是启发式的,缺乏理论基础。我们对现有神经元的紧密表征进行了彻底的经验分析,并揭示它们仅在特定的神经网络上是优越的。然后,我们将网络紧密度的概念介绍为统一的紧密度定义,并表明计算网络紧密度是一个复杂的非convex优化问题。我们通过两个有效的,最紧密的近似值从不同的角度绕过复杂性。结果表明,我们在艺术状态下的方法实现了有希望的表现:(i)达到高达251.28%的改善,以提高认证的较低鲁棒性界限; (ii)在卷积网络上表现出更为精确的验证结果。
translated by 谷歌翻译
我们有助于更好地理解由具有Relu激活和给定架构的神经网络表示的功能。使用来自混合整数优化,多面体理论和热带几何的技术,我们为普遍近似定理提供了数学逆向,这表明单个隐藏层足以用于学习任务。特别是,我们调查完全可增值功能是否完全可以通过添加更多层(没有限制大小)来严格增加。由于它为神经假设类别代表的函数类提供给算法和统计方面,这个问题对算法和统计方面具有潜在的影响。然而,据我们所知,这个问题尚未在神经网络文学中调查。我们还在这些神经假设类别中代表功能所需的神经网络的大小上存在上限。
translated by 谷歌翻译
由于其在输入空间子集上的功能的知识,因此可以根据情况,诅咒或祝福来恢复神经网络的参数权重和偏差的可能性。一方面,恢复参数允许更好的对抗攻击,并且还可以从用于构造网络的数据集中披露敏感信息。另一方面,如果可以恢复网络的参数,它可以保证用户可以解释潜在空间中的特征。它还提供基础,以获得对网络性能的正式保障。因此,表征可以识别其参数的网络以及其参数不能的网络是很重要的。在本文中,我们在深度全连接的前馈recu网络上提供了一组条件,在该馈电中,网络的参数是唯一识别的模型置换和正重型 - 从其实现输入空间的子集。
translated by 谷歌翻译
We consider the algorithmic problem of finding the optimal weights and biases for a two-layer fully connected neural network to fit a given set of data points. This problem is known as empirical risk minimization in the machine learning community. We show that the problem is $\exists\mathbb{R}$-complete. This complexity class can be defined as the set of algorithmic problems that are polynomial-time equivalent to finding real roots of a polynomial with integer coefficients. Furthermore, we show that arbitrary algebraic numbers are required as weights to be able to train some instances to optimality, even if all data points are rational. Our results hold even if the following restrictions are all added simultaneously. $\bullet$ There are exactly two output neurons. $\bullet$ There are exactly two input neurons. $\bullet$ The data has only 13 different labels. $\bullet$ The number of hidden neurons is a constant fraction of the number of data points. $\bullet$ The target training error is zero. $\bullet$ The ReLU activation function is used. This shows that even very simple networks are difficult to train. The result explains why typical methods for $\mathsf{NP}$-complete problems, like mixed-integer programming or SAT-solving, cannot train neural networks to global optimality, unless $\mathsf{NP}=\exists\mathbb{R}$. We strengthen a recent result by Abrahamsen, Kleist and Miltzow [NeurIPS 2021].
translated by 谷歌翻译
Existing neural network verifiers compute a proof that each input is handled correctly under a given perturbation by propagating a symbolic abstraction of reachable values at each layer. This process is repeated from scratch independently for each input (e.g., image) and perturbation (e.g., rotation), leading to an expensive overall proof effort when handling an entire dataset. In this work, we introduce a new method for reducing this verification cost without losing precision based on a key insight that abstractions obtained at intermediate layers for different inputs and perturbations can overlap or contain each other. Leveraging our insight, we introduce the general concept of shared certificates, enabling proof effort reuse across multiple inputs to reduce overall verification costs. We perform an extensive experimental evaluation to demonstrate the effectiveness of shared certificates in reducing the verification cost on a range of datasets and attack specifications on image classifiers including the popular patch and geometric perturbations. We release our implementation at https://github.com/eth-sri/proof-sharing.
translated by 谷歌翻译
神经网络已广泛应用于垃圾邮件和网络钓鱼检测,入侵预防和恶意软件检测等安全应用程序。但是,这种黑盒方法通常在应用中具有不确定性和不良的解释性。此外,神经网络本身通常容易受到对抗攻击的影响。由于这些原因,人们对可信赖和严格的方法有很高的需求来验证神经网络模型的鲁棒性。对抗性的鲁棒性在处理恶意操纵输入时涉及神经网络的可靠性,是安全和机器学习中最热门的主题之一。在这项工作中,我们在神经网络的对抗性鲁棒性验证中调查了现有文献,并在机器学习,安全和软件工程领域收集了39项多元化研究工作。我们系统地分析了它们的方法,包括如何制定鲁棒性,使用哪种验证技术以及每种技术的优势和局限性。我们从正式验证的角度提供分类学,以全面理解该主题。我们根据财产规范,减少问题和推理策略对现有技术进行分类。我们还展示了使用样本模型在现有研究中应用的代表性技术。最后,我们讨论了未来研究的开放问题。
translated by 谷歌翻译
深度分离结果提出了对深度神经网络过较浅的架构的好处的理论解释,建立前者具有卓越的近似能力。然而,没有已知的结果,其中更深的架构利用这种优势成为可提供的优化保证。我们证明,当数据由具有满足某些温和假设的径向对称的分布产生的数据时,梯度下降可以使用具有两层S形激活的深度2神经网络有效地学习球指示器功能,并且隐藏层固定在一起训练。由于众所周知,当使用用单层非线性的深度2网络(Safran和Shamir,2017)使用深度2网络时,球指示器难以近似于一定的重型分配,这建立了我们最好的知识,基于第一优化的分离结果,其中近似架构的近似效益在实践中可怕的。我们的证明技术依赖于随机特征方法,该方法减少了用单个神经元学习的问题,其中新工具需要在数据分布重尾时显示梯度下降的收敛。
translated by 谷歌翻译
我们调查(深)神经网络的可达性问题的复杂性:它是否计算了一些有效输入的有效输出?最近声称问题是NP-Complete,用于一般神经网络和联合输入/输出规范。我们修复了原始上限和下限的缺陷。然后,我们表明NP-HATDINGS已经持有限制的简单规范和具有一层的神经网络,以及针对发生参数的最小要求的神经网络。
translated by 谷歌翻译
While deep neural networks (DNNs) have demonstrated impressive performance in solving many challenging tasks, they are limited to resource-constrained devices owing to their demand for computation power and storage space. Quantization is one of the most promising techniques to address this issue by quantizing the weights and/or activation tensors of a DNN into lower bit-width fixed-point numbers. While quantization has been empirically shown to introduce minor accuracy loss, it lacks formal guarantees on that, especially when the resulting quantized neural networks (QNNs) are deployed in safety-critical applications. A majority of existing verification methods focus exclusively on individual neural networks, either DNNs or QNNs. While promising attempts have been made to verify the quantization error bound between DNNs and their quantized counterparts, they are not complete and more importantly do not support fully quantified neural networks, namely, only weights are quantized. To fill this gap, in this work, we propose a quantization error bound verification method (QEBVerif), where both weights and activation tensors are quantized. QEBVerif consists of two analyses: a differential reachability analysis (DRA) and a mixed-integer linear programming (MILP) based verification method. DRA performs difference analysis between the DNN and its quantized counterpart layer-by-layer to efficiently compute a tight quantization error interval. If it fails to prove the error bound, then we encode the verification problem into an equivalent MILP problem which can be solved by off-the-shelf solvers. Thus, QEBVerif is sound, complete, and arguably efficient. We implement QEBVerif in a tool and conduct extensive experiments, showing its effectiveness and efficiency.
translated by 谷歌翻译
本文通过引入几何深度学习(GDL)框架来构建通用馈电型型模型与可区分的流形几何形状兼容的通用馈电型模型,从而解决了对非欧国人数据进行处理的需求。我们表明,我们的GDL模型可以在受控最大直径的紧凑型组上均匀地近似任何连续目标函数。我们在近似GDL模型的深度上获得了最大直径和上限的曲率依赖性下限。相反,我们发现任何两个非分类紧凑型歧管之间始终都有连续的函数,任何“局部定义”的GDL模型都不能均匀地近似。我们的最后一个主要结果确定了数据依赖性条件,确保实施我们近似的GDL模型破坏了“维度的诅咒”。我们发现,任何“现实世界”(即有限)数据集始终满足我们的状况,相反,如果目标函数平滑,则任何数据集都满足我们的要求。作为应用,我们确认了以下GDL模型的通用近似功能:Ganea等。 (2018)的双波利馈电网络,实施Krishnan等人的体系结构。 (2015年)的深卡尔曼 - 滤波器和深度玛克斯分类器。我们构建了:Meyer等人的SPD-Matrix回归剂的通用扩展/变体。 (2011)和Fletcher(2003)的Procrustean回归剂。在欧几里得的环境中,我们的结果暗示了Kidger和Lyons(2020)的近似定理和Yarotsky和Zhevnerchuk(2019)无估计近似率的数据依赖性版本的定量版本。
translated by 谷歌翻译
我们考虑使用对抗鲁棒性学习的样本复杂性。对于此问题的大多数现有理论结果已经考虑了数据中不同类别在一起或重叠的设置。通过一些实际应用程序,我们认为,相比之下,存在具有完美精度和稳健性的分类器的分类器的良好分离的情况,并表明样品复杂性叙述了一个完全不同的故事。具体地,对于线性分类器,我们显示了大类分离的分布式,其中任何算法的预期鲁棒丢失至少是$ \ω(\ FRAC {D} {n})$,而最大边距算法已预期标准亏损$ o(\ frac {1} {n})$。这表明了通过现有技术不能获得的标准和鲁棒损耗中的间隙。另外,我们介绍了一种算法,给定鲁棒率半径远小于类之间的间隙的实例,给出了预期鲁棒损失的解决方案是$ O(\ FRAC {1} {n})$。这表明,对于非常好的数据,可实现$ O(\ FRAC {1} {n})$的收敛速度,否则就是这样。我们的结果适用于任何$ \ ell_p $ norm以$ p> 1 $(包括$ p = \ idty $)为稳健。
translated by 谷歌翻译
There has been a rapid development and interest in adversarial training and defenses in the machine learning community in the recent years. One line of research focuses on improving the performance and efficiency of adversarial robustness certificates for neural networks \cite{gowal:19, wong_zico:18, raghunathan:18, WengTowardsFC:18, wong:scalable:18, singh:convex_barrier:19, Huang_etal:19, single-neuron-relax:20, Zhang2020TowardsSA}. While each providing a certification to lower (or upper) bound the true distortion under adversarial attacks via relaxation, less studied was the tightness of relaxation. In this paper, we analyze a family of linear outer approximation based certificate methods via a meta algorithm, IBP-Lin. The aforementioned works often lack quantitative analysis to answer questions such as how does the performance of the certificate method depend on the network configuration and the choice of approximation parameters. Under our framework, we make a first attempt at answering these questions, which reveals that the tightness of linear approximation based certification can depend heavily on the configuration of the trained networks.
translated by 谷歌翻译
在本文中,我们在具有线性阈值激活功能的神经网络上提出了新的结果。我们精确地表征了这种神经网络可表示的功能,并且显示2个隐藏层是必要的并且足以表示类中可表示的任何功能。鉴于使用其他流行的激活功能的神经网络的最近精确的可比性调查,这是一个令人惊讶的结果,这些功能使用其他流行的激活功能,如整流的线性单元(Relu)。我们还给出了代表类中任意函数所需的神经网络的大小的精确界限。最后,我们设计了一种算法来解决具有固定架构的这些神经网络的全球最优性的经验风险最小化(ERM)问题。如果输入维度和网络架构的大小被认为是固定常数,则算法的运行时间是数据样本大小的多项式。该算法的意义上是独一无二的,即它适用于任何数量的层数,而先前的多项式时间全局最佳算法仅适用于非常受限制的架构类。
translated by 谷歌翻译
我们研究了神经网络中平方损耗训练问题的优化景观和稳定性,但通用非线性圆锥近似方案。据证明,如果认为非线性圆锥近似方案是(以适当定义的意义)比经典线性近似方法更具表现力,并且如果存在不完美的标签向量,则在方位损耗的训练问题必须在其中不稳定感知其解决方案集在训练数据中的标签向量上不连续地取决于标签向量。我们进一步证明对这些不稳定属性负责的效果也是马鞍点出现的原因和杂散的局部最小值,这可能是从全球解决方案的任意遥远的,并且既不训练问题也不是训练问题的不稳定性通常,杂散局部最小值的存在可以通过向目标函数添加正则化术语来克服衡量近似方案中参数大小的目标函数。无论可实现的可实现性是否满足,后一种结果都被证明是正确的。我们表明,我们的分析特别适用于具有可变宽度的自由结插值方案和深层和浅层神经网络的培训问题,其涉及各种激活功能的任意混合(例如,二进制,六骨,Tanh,arctan,软标志, ISRU,Soft-Clip,SQNL,Relu,Lifley Relu,Soft-Plus,Bent Identity,Silu,Isrlu和ELU)。总之,本文的发现说明了神经网络和一般非线性圆锥近似仪器的改进近似特性以直接和可量化的方式与必须解决的优化问题的不期望的性质链接,以便训练它们。
translated by 谷歌翻译
我们研究神经网络表达能力的基本限制。给定两组$ f $,$ g $的实值函数,我们首先证明了$ f $中的功能的一般下限,可以在$ l^p(\ mu)$ norm中通过$ g中的功能近似$,对于任何$ p \ geq 1 $和任何概率度量$ \ mu $。下限取决于$ f $的包装数,$ f $的范围以及$ g $的脂肪震动尺寸。然后,我们实例化了$ g $对应于分段的馈电神经网络的情况,并详细描述了两组$ f $:h {\“ o} lder balls和多变量单调函数。除了匹配(已知或新的)上限与日志因素外,我们的下限还阐明了$ l^p $ Norm或SUP Norm中近似之间的相似性或差异,解决了Devore等人的开放问题(2021年))。我们的证明策略与SUP Norm案例不同,并使用了Mendelson(2002)的关键概率结果。
translated by 谷歌翻译
Neural networks are increasingly applied in safety critical domains, their verification thus is gaining importance. A large class of recent algorithms for proving input-output relations of feed-forward neural networks are based on linear relaxations and symbolic interval propagation. However, due to variable dependencies, the approximations deteriorate with increasing depth of the network. In this paper we present DPNeurifyFV, a novel branch-and-bound solver for ReLU networks with low dimensional input-space that is based on symbolic interval propagation with fresh variables and input-splitting. A new heuristic for choosing the fresh variables allows to ameliorate the dependency problem, while our novel splitting heuristic, in combination with several other improvements, speeds up the branch-and-bound procedure. We evaluate our approach on the airborne collision avoidance networks ACAS Xu and demonstrate runtime improvements compared to state-of-the-art tools.
translated by 谷歌翻译