Deep learning has achieved remarkable success in numerous domains with help from large
amounts of big data. However, the quality of data labels is a concern because of the lack of …

The fault diagnosis (FD) of rolling bearing (RB) has a great significance in safe operation of
engineering equipment. Many intelligent diagnosis methods have been successfully …

Existing research on learning with noisy labels mainly focuses on synthetic label noise.
Synthetic noise, though has clean structures which greatly enabled statistical analyses, often …

Recently, over-parameterized deep networks, with increasingly more network parameters
than training samples, have dominated the performances of modern machine learning …

Label smoothing is commonly used in training deep learning models, wherein one-hot
training labels are mixed with uniform label vectors. Empirically, smoothing has been shown …

Human-annotated labels are often prone to noise, and the presence of such noise will
degrade the performance of the resulting deep neural network (DNN) models. Much of the …

Self-supervised contrastive representation learning has proved incredibly successful in the
vision and natural language domains, enabling state-of-the-art performance with orders of …

Because learning sometimes involves sensitive data, machine learning algorithms have
been extended to offer differential privacy for training data. In practice, this has been mostly …

In this work, we study the large-scale pretraining of BERT-Large with differentially private
SGD (DP-SGD). We show that combined with a careful implementation, scaling up the batch …

The presence of label noise often misleads the training of deep neural networks. Departing
from the recent literature which largely assumes the label noise rate is only determined by …