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
Verifying the robustness property of a general Rectified Linear Unit (ReLU) network is an NPcomplete problem. Although finding the exact minimum adversarial distortion is hard, giving a certified lower bound of the minimum distortion is possible. Current available methods of computing such a bound are either time-consuming or deliver low quality bounds that are too loose to be useful. In this paper, we exploit the special structure of ReLU networks and provide two computationally efficient algorithms (Fast-Lin,Fast-Lip) that are able to certify non-trivial lower bounds of minimum adversarial distortions. Experiments show that (1) our methods deliver bounds close to (the gap is 2-3X) exact minimum distortions found by Reluplex in small networks while our algorithms are more than 10,000 times faster; (2) our methods deliver similar quality of bounds (the gap is within 35% and usually around 10%; sometimes our bounds are even better) for larger networks compared to the methods based on solving linear programming problems but our algorithms are 33-14,000 times faster; (3) our method is capable of solving large MNIST and CIFAR networks up to 7 layers with more than 10,000 neurons within tens of seconds on a single CPU core. In addition, we show that there is no polynomial time algorithm that can approximately find the minimum 1 adversarial distortion of a ReLU network with a 0.99 ln n approximation ratio unless NP=P, where n is the number of neurons in the network.
translated by 谷歌翻译
随着深度学习在关键任务系统中的越来越多的应用,越来越需要对神经网络的行为进行正式保证。确实,最近提出了许多用于验证神经网络的方法,但是这些方法通常以有限的可伸缩性或不足的精度而挣扎。许多最先进的验证方案中的关键组成部分是在网络中可以为特定输入域获得的神经元获得的值计算下限和上限 - 并且这些界限更紧密,验证的可能性越大,验证的可能性就越大。成功。计算这些边界的许多常见算法是符号结合传播方法的变化。其中,利用一种称为后替代的过程的方法特别成功。在本文中,我们提出了一种使背部替代产生更严格的界限的方法。为了实现这一目标,我们制定并最大程度地减少背部固定过程中发生的不精确错误。我们的技术是一般的,从某种意义上说,它可以将其集成到许多现有的符号结合的传播技术中,并且只有较小的修改。我们将方法作为概念验证工具实施,并且与执行背部替代的最先进的验证者相比,取得了有利的结果。
translated by 谷歌翻译
To rigorously certify the robustness of neural networks to adversarial perturbations, most state-of-the-art techniques rely on a triangle-shaped linear programming (LP) relaxation of the ReLU activation. While the LP relaxation is exact for a single neuron, recent results suggest that it faces an inherent "convex relaxation barrier" as additional activations are added, and as the attack budget is increased. In this paper, we propose a nonconvex relaxation for the ReLU relaxation, based on a low-rank restriction of a semidefinite programming (SDP) relaxation. We show that the nonconvex relaxation has a similar complexity to the LP relaxation, but enjoys improved tightness that is comparable to the much more expensive SDP relaxation. Despite nonconvexity, we prove that the verification problem satisfies constraint qualification, and therefore a Riemannian staircase approach is guaranteed to compute a near-globally optimal solution in polynomial time. Our experiments provide evidence that our nonconvex relaxation almost completely overcome the "convex relaxation barrier" faced by the LP relaxation.
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
基于基于不完整的神经网络验证如冠的绑定传播非常有效,可以显着加速基于神经网络的分支和绑定(BAB)。然而,绑定的传播不能完全处理由昂贵的线性编程(LP)求解器的BAB常规引入的神经元分割限制,导致界限和损伤验证效率。在这项工作中,我们开发了一种基于$ \ beta $ -cra所做的,一种基于新的绑定传播方法,可以通过从原始或双空间构造的可优化参数$ \ beta $完全编码神经元分割。当在中间层中联合优化时,$ \ Beta $ -CROWN通常会产生比具有神经元分裂约束的典型LP验证更好的界限,同时像GPU上的皇冠一样高效且并行化。适用于完全稳健的验证基准,使用BAB的$ \ Beta $ -CROWN比基于LP的BAB方法快三个数量级,并且比所有现有方法更快,同时产生较低的超时率。通过早期终止BAB,我们的方法也可用于有效的不完整验证。与强大的不完整验证者相比,我们始终如一地在许多设置中获得更高的验证准确性,包括基于凸屏障破碎技术的验证技术。与最严重但非常昂贵的Semidefinite编程(SDP)的不完整验证者相比,我们获得了更高的验证精度,验证时间较少三个级。我们的算法授权$ \ alpha,\ \β$ -craft(Alpha-Beta-Crown)验证者,VNN-Comp 2021中的获胜工具。我们的代码可在http://papercode.cc/betacrown提供
translated by 谷歌翻译
Despite their impressive performance on diverse tasks, neural networks fail catastrophically in the presence of adversarial inputs-imperceptibly but adversarially perturbed versions of natural inputs. We have witnessed an arms race between defenders who attempt to train robust networks and attackers who try to construct adversarial examples. One promise of ending the arms race is developing certified defenses, ones which are provably robust against all attackers in some family. These certified defenses are based on convex relaxations which construct an upper bound on the worst case loss over all attackers in the family. Previous relaxations are loose on networks that are not trained against the respective relaxation. In this paper, we propose a new semidefinite relaxation for certifying robustness that applies to arbitrary ReLU networks. We show that our proposed relaxation is tighter than previous relaxations and produces meaningful robustness guarantees on three different foreign networks whose training objectives are agnostic to our proposed relaxation.32nd Conference on Neural Information Processing Systems (NIPS 2018),
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
We propose a method to learn deep ReLU-based classifiers that are provably robust against normbounded adversarial perturbations on the training data. For previously unseen examples, the approach is guaranteed to detect all adversarial examples, though it may flag some non-adversarial examples as well. The basic idea is to consider a convex outer approximation of the set of activations reachable through a norm-bounded perturbation, and we develop a robust optimization procedure that minimizes the worst case loss over this outer region (via a linear program). Crucially, we show that the dual problem to this linear program can be represented itself as a deep network similar to the backpropagation network, leading to very efficient optimization approaches that produce guaranteed bounds on the robust loss. The end result is that by executing a few more forward and backward passes through a slightly modified version of the original network (though possibly with much larger batch sizes), we can learn a classifier that is provably robust to any norm-bounded adversarial attack. We illustrate the approach on a number of tasks to train classifiers with robust adversarial guarantees (e.g. for MNIST, we produce a convolutional classifier that provably has less than 5.8% test error for any adversarial attack with bounded ∞ norm less than = 0.1), and code for all experiments is available at http://github.com/ locuslab/convex_adversarial.
translated by 谷歌翻译
人工神经网络(ANN)训练景观的非凸起带来了固有的优化困难。虽然传统的背传播随机梯度下降(SGD)算法及其变体在某些情况下是有效的,但它们可以陷入杂散的局部最小值,并且对初始化和普通公共表敏感。最近的工作表明,随着Relu激活的ANN的培训可以重新重整为凸面计划,使希望能够全局优化可解释的ANN。然而,天真地解决凸训练制剂具有指数复杂性,甚至近似启发式需要立方时间。在这项工作中,我们描述了这种近似的质量,并开发了两个有效的算法,这些算法通过全球收敛保证培训。第一算法基于乘法器(ADMM)的交替方向方法。它解决了精确的凸形配方和近似对应物。实现线性全局收敛,并且初始几次迭代通常会产生具有高预测精度的解决方案。求解近似配方时,每次迭代时间复杂度是二次的。基于“采样凸面”理论的第二种算法更简单地实现。它解决了不受约束的凸形制剂,并收敛到大约全球最佳的分类器。当考虑对抗性培训时,ANN训练景观的非凸起加剧了。我们将稳健的凸优化理论应用于凸训练,开发凸起的凸起制剂,培训Anns对抗对抗投入。我们的分析明确地关注一个隐藏层完全连接的ANN,但可以扩展到更复杂的体系结构。
translated by 谷歌翻译
分析深神经网络对输入扰动的最坏情况的性能等于解决一个大规模的非凸优化问题,过去的几项工作提出了凸出的放松作为有希望的替代方案。但是,即使对于合理的神经网络,这些放松也无法处理,因此必须在实践中被较弱的放松所取代。在这项工作中,我们提出了一种新型的操作员分裂方法,该方法可以将问题直接解决至高精度的凸松弛,从而将其拆分为经常具有分析溶液的较小的子问题。该方法是模块化的,范围为非常大的问题实例,并损害了与GPU加速的快速并行化的操作。我们展示了我们在图像分类和强化学习设置以及神经网络动力学系统的可及性分析中界定大型卷积网络最差的方法的方法。
translated by 谷歌翻译
We present a new algorithm for automatically bounding the Taylor remainder series. In the special case of a scalar function $f: \mathbb{R} \mapsto \mathbb{R}$, our algorithm takes as input a reference point $x_0$, trust region $[a, b]$, and integer $k \ge 0$, and returns an interval $I$ such that $f(x) - \sum_{i=0}^k \frac {f^{(i)}(x_0)} {i!} (x - x_0)^i \in I (x - x_0)^{k+1}$ for all $x \in [a, b]$. As in automatic differentiation, the function $f$ is provided to the algorithm in symbolic form, and must be composed of known elementary functions. At a high level, our algorithm has two steps. First, for a variety of commonly-used elementary functions (e.g., $\exp$, $\log$), we derive sharp polynomial upper and lower bounds on the Taylor remainder series. We then recursively combine the bounds for the elementary functions using an interval arithmetic variant of Taylor-mode automatic differentiation. Our algorithm can make efficient use of machine learning hardware accelerators, and we provide an open source implementation in JAX. We then turn our attention to applications. Most notably, we use our new machinery to create the first universal majorization-minimization optimization algorithms: algorithms that iteratively minimize an arbitrary loss using a majorizer that is derived automatically, rather than by hand. Applied to machine learning, this leads to architecture-specific optimizers for training deep networks that converge from any starting point, without hyperparameter tuning. Our experiments show that for some optimization problems, these hyperparameter-free optimizers outperform tuned versions of gradient descent, Adam, and AdaGrad. We also show that our automatically-derived bounds can be used for verified global optimization and numerical integration, and to prove sharper versions of Jensen's inequality.
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
训练深神网络是一个众所周知的高度非凸问题。在最近的作品中,显示出具有RELU激活的正则化两层神经网络没有二元性差距,这可以通过凸面程序进行全局优化。对于具有向量输出的多层线性网络,我们提出了凸双问题,并证明对偶性差距对于深度三和更深的网络而言并非零。但是,通过将深层网络修改为更强大的并行体系结构,我们表明二元性差距完全为零。因此,强大的凸面双重性具有,因此存在等效的凸面程序,使培训深层网络达到全球最优性。我们还证明,参数中的重量衰减正则化明确地通过封闭形式表达式鼓励低级溶液。对于三层非平行relu网络,我们表明对级别1数据矩阵的强双重性具有强度,但是,对白色数据矩阵的二元性差距不是零。同样,通过将神经网络体系结构转换为相应的并行版本,二元性差距消失了。
translated by 谷歌翻译
在负面的感知问题中,我们给出了$ n $数据点$({\ boldsymbol x} _i,y_i)$,其中$ {\ boldsymbol x} _i $是$ d $ -densional vector和$ y_i \ in \ { + 1,-1 \} $是二进制标签。数据不是线性可分离的,因此我们满足自己的内容,以找到最大的线性分类器,具有最大的\ emph {否定}余量。换句话说,我们想找到一个单位常规矢量$ {\ boldsymbol \ theta} $,最大化$ \ min_ {i \ le n} y_i \ langle {\ boldsymbol \ theta},{\ boldsymbol x} _i \ rangle $ 。这是一个非凸优化问题(它相当于在Polytope中找到最大标准矢量),我们在两个随机模型下研究其典型属性。我们考虑比例渐近,其中$ n,d \ to \ idty $以$ n / d \ to \ delta $,并在最大边缘$ \ kappa _ {\ text {s}}(\ delta)上证明了上限和下限)$或 - 等效 - 在其逆函数$ \ delta _ {\ text {s}}(\ kappa)$。换句话说,$ \ delta _ {\ text {s}}(\ kappa)$是overparametization阈值:以$ n / d \ le \ delta _ {\ text {s}}(\ kappa) - \ varepsilon $一个分类器实现了消失的训练错误,具有高概率,而以$ n / d \ ge \ delta _ {\ text {s}}(\ kappa)+ \ varepsilon $。我们在$ \ delta _ {\ text {s}}(\ kappa)$匹配,以$ \ kappa \ to - \ idty $匹配。然后,我们分析了线性编程算法来查找解决方案,并表征相应的阈值$ \ delta _ {\ text {lin}}(\ kappa)$。我们观察插值阈值$ \ delta _ {\ text {s}}(\ kappa)$和线性编程阈值$ \ delta _ {\ text {lin {lin}}(\ kappa)$之间的差距,提出了行为的问题其他算法。
translated by 谷歌翻译
我们提供匹配的Under $ \ sigma ^ 2 / \ log(d / n)$的匹配的上下界限为最低$ \ ell_1 $ -norm插值器,a.k.a.基础追踪。我们的结果紧紧达到可忽略的术语,而且是第一个暗示噪声最小范围内插值的渐近一致性,因为各向同性特征和稀疏的地面真理。我们的工作对最低$ \ ell_2 $ -norm插值的“良性接收”进行了补充文献,其中才能在特征有效地低维时实现渐近一致性。
translated by 谷歌翻译
在本文中,我们研究了使用深丽升方法(DRM)和物理信息的神经网络(Pinns)从随机样品求解椭圆局部微分方程(PDE)的深度学习技术的统计限制。为了简化问题,我们专注于原型椭圆PDE:SCHR \“odinginger方程,具有零的Dirichlet边界条件,其在量子 - 机械系统中具有广泛的应用。我们为两种方法建立了上下界,通过快速速率泛化绑定并发地改善了这个问题的上限。我们发现当前的深ritz方法是次优的,提出修改版本。我们还证明了Pinn和DRM的修改版本可以实现Minimax SoboLev空间的最佳限制。经验上,近期工作表明,根据权力法,我们提供了培训训练的深层模型精度,我们提供了计算实验,以显示对深PDE求解器的尺寸依赖权力法的类似行为。
translated by 谷歌翻译
经认证的稳健性是安全关键应用中的深度神经网络的理想性质,流行的训练算法可以通过计算其Lipschitz常数的全球界限来认证神经网络的鲁棒性。然而,这种界限往往松动:它倾向于过度规范神经网络并降低其自然精度。绑定的Lipschitz绑定可以在自然和认证的准确性之间提供更好的权衡,但通常很难根据网络的非凸起计算。在这项工作中,我们通过考虑激活函数(例如Relu)和权重矩阵之间的相互作用,提出了一种有效和培训的\ emph {本地} Lipschitz上限。具体地,当计算权重矩阵的诱发标准时,我们消除了相应的行和列,其中保证激活函数在每个给定数据点的邻域中是常数,它提供比全局Lipschitz常数的可怕更严格的绑定神经网络。我们的方法可用作插入式模块,以拧紧在许多可认证的训练算法中绑定的Lipschitz。此外,我们建议夹住激活功能(例如,Relu和Maxmin),具有可读的上限阈值和稀疏性损失,以帮助网络实现甚至更严格的本地嘴唇尖端。在实验上,我们表明我们的方法始终如一地优于Mnist,CiFar-10和Tinyimagenet数据集的清洁和认证准确性,具有各种网络架构的清洁和认证的准确性。
translated by 谷歌翻译
神经网络(NNS)已成功地用于代表复杂动力学系统的状态演变。这样的模型,称为NN动态模型(NNDMS),使用NN的迭代噪声预测来估计随时间推移系统轨迹的分布。尽管它们的准确性,但对NNDMS的安全分析仍然是一个具有挑战性的问题,并且在很大程度上尚未探索。为了解决这个问题,在本文中,我们介绍了一种为NNDM提供安全保证的方法。我们的方法基于随机屏障函数,其与安全性的关系类似于Lyapunov功能的稳定性。我们首先展示了通过凸优化问题合成NNDMS随机屏障函数的方法,该问题又为系统的安全概率提供了下限。我们方法中的一个关键步骤是,NNS的最新凸近似结果的利用是找到零件线性边界,这允许将屏障函数合成问题作为一个方形优化程序的制定。如果获得的安全概率高于所需的阈值,则该系统将获得认证。否则,我们引入了一种生成控制系统的方法,该系统以最小的侵入性方式稳健地最大化安全概率。我们利用屏障函数的凸属性来提出最佳控制合成问题作为线性程序。实验结果说明了该方法的功效。即,他们表明该方法可以扩展到具有多层和数百个神经元的多维NNDM,并且控制器可以显着提高安全性概率。
translated by 谷歌翻译
最近的一项工作已经通过神经切线核(NTK)分析了深神经网络的理论特性。特别是,NTK的最小特征值与记忆能力,梯度下降算法的全球收敛性和深网的概括有关。但是,现有结果要么在两层设置中提供边界,要么假设对于多层网络,将NTK矩阵的频谱从0界限为界限。在本文中,我们在无限宽度和有限宽度的限制情况下,在最小的ntk矩阵的最小特征值上提供了紧密的界限。在有限宽度的设置中,我们认为的网络体系结构相当笼统:我们需要大致订购$ n $神经元的宽层,$ n $是数据示例的数量;剩余层宽度的缩放是任意的(取决于对数因素)。为了获得我们的结果,我们分析了各种量的独立兴趣:我们对隐藏特征矩阵的最小奇异值以及输入输出特征图的Lipschitz常数上的上限给出了下限。
translated by 谷歌翻译