Affiliations: [a] Office of Research, University of the Sunshine Coast, QLD, Maroochydore, Australia | [b] School of Science, Information Technology and Engineering, University of Ballarat, Ballarat, VIC, Australia | [c] National ICT Australia, VRL, Melbourne, VIC, Australia
Correspondence:
[*]
Corresponding author: Lei Zhao, Research Information Coordinator, Office of Research, University of the Sunshine Coast, Maroochydore, DC QLD 4558, Australia. E-mail: [email protected].
Abstract: Loss function plays an important role in data classification. Manyloss functions have been proposed and applied to differentclassification problems. This paper proposes a new so called thesmoothed 0-1 loss function, that could be considered as anapproximation of the classical 0-1 loss function. Due to thenon-convexity property of the proposed loss function, globaloptimization methods are required to solve the correspondingoptimization problems. Together with the proposed loss function, wecompare the performance of several existing loss functions in theclassification of noisy data sets. In this comparison, differentoptimization problems are considered in regards to the convexity andsmoothness of different loss functions. The experimental resultsshow that the proposed smoothed 0-1 loss function works better ondata sets with noisy labels, noisy features, and outliers.
Keywords: Classification, loss function, machine learning, optimization, data mining
