While cross entropy (CE) is the most commonly used loss function to train deep neural
networks for classification tasks, many alternative losses have been developed to obtain …

Previous work has proposed many new loss functions and regularizers that improve test
accuracy on image classification tasks. However, it is not clear whether these loss functions …

A fundamental challenge for machine learning models is how to generalize learned models
for out-of-distribution (OOD) data. Among various approaches, exploiting invariant features …

Retrogressive thaw slumps (RTS) are thermokarst features in ice-rich hillslope permafrost
terrain, and their occurrence in the warming Arctic is increasingly frequent and has caused …

Road damage detection is very important for road safety and timely repair. The previous
detection methods mainly rely on humans or large machines, which are costly and …

Label smoothing (LS) is an arising learning paradigm that uses the positively weighted
average of both the hard training labels and uniformly distributed soft labels. It was shown …

Large-scale, high-quality data are considered an essential factor for the successful
application of many deep learning techniques. Meanwhile, numerous real-world deep …

Label Smoothing (LS) is a widely-used training technique that adjusts hard labels towards a
softer distribution, which prevents model being over-confidence and enhances model …

Abstract The one-hot 0/1 encoding method is the most popularized encoding method of
class labels for classification tasks. Despite its simplicity and popularity, it comes with …

Label smoothing is a widely used technique in various domains, such as image
classification and speech recognition, known for effectively combating model overfitting …