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 2010.07532cs.LGmath.OCstat.ML
收藏

从样本证明神经网络对随机输入噪声的鲁棒性

Certifying Neural Network Robustness to Random Input Noise from Samples

Brendon G. Anderson, Somayeh Sojoudi

摘要:

证明神经网络在输入不确定性存在下的鲁棒性的方法在安全关键设置中是至关重要的。文献中的大多数认证方法都是针对对抗性输入不确定性而设计的,但研究人员最近表明需要考虑随机不确定性的方法。在本文中,我们提出了一种新的鲁棒性认证方法,上限的错误分类的概率时,输入噪声遵循任意概率分布。该界被转换为机会约束优化问题,然后使用输入输出样本来替换优化约束。由此产生的优化减少到一个线性规划的解析解。此外,我们开发了一个充分条件,所需的样本数量,使误分类界举行压倒性的概率。我们对MNIST分类器的案例研究表明,该方法能够证明一个均匀的无穷范数不确定区域,其半径比目前最先进的方法可以证明的半径大近50倍。

英文摘要:

Methods to certify the robustness of neural networks in the presence of input uncertainty are vital in safety-critical settings. Most certification methods in the literature are designed for adversarial input uncertainty, but researchers have recently shown a need for methods that consider random uncertainty. In this paper, we propose a novel robustness certification method that upper bounds the probability of misclassification when the input noise follows an arbitrary probability distribution. This bound is cast as a chance-constrained optimization problem, which is then reformulated using input-output samples to replace the optimization constraints. The resulting optimization reduces to a linear program with an analytical solution. Furthermore, we develop a sufficient condition on the number of samples needed to make the misclassification bound hold with overwhelming probability. Our case studies on MNIST classifiers show that this method is able to certify a uniform infinity-norm uncertainty region with a radius of nearly 50 times larger than what the current state-of-the-art method can certify.

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。

↑