The instance discrimination paradigm has become dominant in unsupervised learning. It
always adopts a teacher-student framework, in which the teacher provides embedded …

Contrastive learning methods for unsupervised visual representation learning have reached
remarkable levels of transfer performance. We argue that the power of contrastive learning …

Self-supervised learning has recently emerged as the preeminent pretraining paradigm
across and between modalities, with remarkable results. In the image domain specifically …

Contrastive learning models based on Siamese structure have demonstrated remarkable
performance in self-supervised learning. Such a success of contrastive learning relies on …

Cluster discrimination is an effective pretext task for unsupervised representation learning,
which often consists of two phases: clustering and discrimination. Clustering is to assign …

Self-supervised learning aims to learn representations from the data itself without explicit
manual supervision. Existing efforts ignore a crucial aspect of self-supervised learning-the …

Unsupervised learning has grown in popularity because of the difficulty of collecting
annotated data and the development of modern frameworks that allow us to learn from …

In this paper, we propose an instance similarity learning (ISL) method for unsupervised
feature representation. Conventional methods assign close instance pairs in the feature …

Self-supervised learning of image representations by predicting future frames is a promising
direction but still remains a challenge. This is because of the under-determined nature of …

Unsupervised representation learning with contrastive learning achieves great success
recently. However, these methods have to duplicate each training batch to construct …