arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2302.07238cs.LGcs.AIcs.NE
收藏

Cauchy损失函数:高斯和Cauchy噪声下的鲁棒性

Cauchy Loss Function: Robustness Under Gaussian and Cauchy Noise

University of Pretoria(比勒陀利亚大学)

Thamsanqa Mlotshwa, Heinrich van Deventer, Anna Sergeevna Bosman

摘要:

在监督机器学习中,损失函数的选择隐含地假设数据上的特定噪声分布。例如,经常使用的均方误差(MSE)损失假设高斯噪声分布。在训练和测试过程中,损失函数的选择会影响人工神经网络(ANN)的性能。众所周知,MSE在存在离群值的情况下可能会产生低于标准的性能。柯西损失函数(CLF)假设柯西噪声分布,因此可能更适合具有离群值的数据。本文旨在确定CLF与MSE相比的稳健性和普遍性程度。CLF和MSE在人工神经网络训练的背景下,在一些手工回归问题和人工模拟离群值的真实回归问题上进行评估。CLF产生的结果与MSE产生的结果相当或更好,但有几个明显的例外。

英文摘要:

In supervised machine learning, the choice of loss function implicitly assumes a particular noise distribution over the data. For example, the frequently used mean squared error (MSE) loss assumes a Gaussian noise distribution. The choice of loss function during training and testing affects the performance of artificial neural networks (ANNs). It is known that MSE may yield substandard performance in the presence of outliers. The Cauchy loss function (CLF) assumes a Cauchy noise distribution, and is therefore potentially better suited for data with outliers. This papers aims to determine the extent of robustness and generalisability of the CLF as compared to MSE. CLF and MSE are assessed on a few handcrafted regression problems, and a real-world regression problem with artificially simulated outliers, in the context of ANN training. CLF yielded results that were either comparable to or better than the results yielded by MSE, with a few notable exceptions.

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。

↑