Curriculum learning (CL) is a training strategy that trains a machine learning model from
easier data to harder data, which imitates the meaningful learning order in human curricula …

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 …

Unsupervised person re-identification (re-ID) aims at learning discriminative representations
for person retrieval from unlabeled data. Recent techniques accomplish this task by using …

Deep networks have strong capacities of embedding data into latent representations and
finishing following tasks. However, the capacities largely come from high-quality annotated …

In this paper, we propose Parametric Contrastive Learning (PaCo) to tackle long-tailed
recognition. Based on theoretical analysis, we observe supervised contrastive loss tends to …

Abstract Noisy label Facial Expression Recognition (FER) is more challenging than
traditional noisy label classification tasks due to the inter-class similarity and the annotation …

In the real open world, data tends to follow long-tailed class distributions, motivating the well-
studied long-tailed recognition (LTR) problem. Naive training produces models that are …

As the class size grows, maintaining a balanced dataset across many classes is challenging
because the data are long-tailed in nature; it is even impossible when the sample-of-interest …

Despite the success of the deep neural networks, it remains challenging to effectively build a
system for long-tail visual recognition tasks. To address this problem, we first investigate the …

Deep classifiers have achieved great success in visual recognition. However, real-world
data is long-tailed by nature, leading to the mismatch between training and testing …