MoCo v2

Introduced by Chen et al. in Improved Baselines with Momentum Contrastive Learning

MoCo v2 is an improved version of the Momentum Contrast self-supervised learning algorithm. Motivated by the findings presented in the SimCLR paper, authors:

Replace the 1-layer fully connected layer with a 2-layer MLP head with ReLU for the unsupervised training stage.
Include blur augmentation.
Use cosine learning rate schedule.

These modifications enable MoCo to outperform the state-of-the-art SimCLR with a smaller batch size and fewer epochs.

Source: Improved Baselines with Momentum Contrastive Learning

Read Paper See Code

Papers

Paper	Code	Results	Date	Stars

Tasks

Task	Papers	Share
Self-Supervised Learning	16	22.54%
Image Classification	7	9.86%
Semantic Segmentation	6	8.45%
Object Detection	5	7.04%
Classification	3	4.23%
Language Modelling	2	2.82%
Instance Segmentation	2	2.82%
Self-Supervised Image Classification	2	2.82%
Clustering	2	2.82%