Sharma Somya, Chatterjee Snigdhansu
Department of Computer Science and Engineering, University of Minnesota-Twin Cities, 200 Union Street SE, Minneapolis, MN 55455, USA.
School of Statistics, University of Minnesota-Twin Cities, 313 Ford Hall, 224 Church St. SE, Minneapolis, MN 55455, USA.
Entropy (Basel). 2021 Nov 20;23(11):1546. doi: 10.3390/e23111546.
With the advent of big data and the popularity of black-box deep learning methods, it is imperative to address the robustness of neural networks to noise and outliers. We propose the use of Winsorization to recover model performances when the data may have outliers and other aberrant observations. We provide a comparative analysis of several probabilistic artificial intelligence and machine learning techniques for supervised learning case studies. Broadly, Winsorization is a versatile technique for accounting for outliers in data. However, different probabilistic machine learning techniques have different levels of efficiency when used on outlier-prone data, with or without Winsorization. We notice that Gaussian processes are extremely vulnerable to outliers, while deep learning techniques in general are more robust.
随着大数据的出现以及黑箱深度学习方法的普及，解决神经网络对噪声和异常值的鲁棒性问题变得势在必行。我们建议在数据可能存在异常值和其他异常观测时使用 Winsor 化来恢复模型性能。我们针对监督学习案例研究对几种概率人工智能和机器学习技术进行了比较分析。广义而言，Winsor 化是一种用于处理数据中异常值的通用技术。然而，不同的概率机器学习技术在处理容易出现异常值的数据时，无论是否使用 Winsor 化，其效率水平都有所不同。我们注意到高斯过程极易受到异常值的影响，而深度学习技术总体上更具鲁棒性。
