{"url":"/method/simclrv2","slug":"simclrv2","name":"SimCLRv2","full_name":"SimCLRv2","full_name_withheld":false,"description_markdown":"**SimCLRv2** is a semi-supervised learning method for learning from few labeled examples while making best use of a large amount of unlabeled data. It is a modification of a recently proposed contrastive learning framework, [SimCLR](https://www.paperswithcode.com/method/simclr). It improves upon it in three major ways:\r\n\r\n1. To fully leverage the power of general pre-training, larger [ResNet](https://paperswithcode.com/method/resnet) models are explored. Unlike SimCLR and other previous work, whose largest model is ResNet-50 (4×), SimCLRv2 trains models that are deeper but less wide. The largest model trained is a 152 layer ResNet with 3× wider channels and [selective kernels](https://paperswithcode.com/method/selective-kernel-convolution) (SK), a channel-wise attention mechanism that improves the parameter efficiency of the network. By scaling up the model from ResNet-50 to ResNet-152 (3×+SK), a 29% relative improvement is obtained in top-1 accuracy when fine-tuned on 1% of labeled examples.\r\n\r\n2. The capacity of the non-linear network $g(·)$ (a.k.a. projection head) is increased, by making it deeper. Furthermore, instead of throwing away $g(·)$ entirely after pre-training as in SimCLR, fine-tuning occurs from a middle layer. This small change yields a significant improvement for both linear evaluation and fine-tuning with only a few labeled examples. Compared to SimCLR with 2-layer projection head, by using a 3-layer projection head and fine-tuning from the 1st layer of projection head, it results in as much as 14% relative improvement in top-1 accuracy when fine-tuned on 1% of labeled examples.\r\n\r\n3. The memory mechanism of [MoCo v2](https://paperswithcode.com/method/moco-v2) is incorporated, which designates a memory network (with a moving average of weights for stabilization) whose output will be buffered as negative examples. Since training is based on large mini-batch which already supplies many contrasting negative examples, this change yields an improvement of ∼1% for linear evaluation as well as when fine-tuning on 1% of labeled examples.","description_state":"present","introduced_year":null,"introduced_by":{"title":"Big Self-Supervised Models are Strong Semi-Supervised Learners","paper":"/paper/big-self-supervised-models-are-strong-semi","first_author":"Ting Chen","n_authors":5,"url_abs":null,"archive_paper_url":"https://paperswithcode.com/paper/big-self-supervised-models-are-strong-semi"},"source":{"url":"https://arxiv.org/abs/2006.10029v2","title":"Big Self-Supervised Models are Strong Semi-Supervised Learners","url_on_a_paper_host":true},"code_snippet_url":null,"code_snippet_url_on_a_code_host":false,"categories":[{"area":"General","area_id":"general","collection":"Semi-Supervised Learning Methods","url":"/methods/category/semi-supervised-learning-methods","pwc_aliases":[]}],"n_papers_tagged":4,"archive_num_papers":4,"papers_newest_first":[{"paper":null,"title":"Differentially Private Optimizers Can Learn Adversarially Robust Models","date":"2022-11-16","arxiv_id":"2211.08942","n_code_links":0,"syntology":null},{"paper":"/paper/weakly-supervised-contrastive-learning-1","title":"Weakly Supervised Contrastive Learning","date":"2021-10-10","arxiv_id":"2110.04770","n_code_links":1,"syntology":{"ran":0,"of":3,"unverified":3,"pointer_only":3}},{"paper":"/paper/can-vision-transformers-learn-without-natural","title":"Can Vision Transformers Learn without Natural Images?","date":"2021-03-24","arxiv_id":"2103.13023","n_code_links":1,"syntology":{"ran":5,"of":11,"unverified":6,"pointer_only":0}},{"paper":"/paper/big-self-supervised-models-are-strong-semi","title":"Big Self-Supervised Models are Strong Semi-Supervised Learners","date":"2020-06-17","arxiv_id":"2006.10029","n_code_links":9,"syntology":{"ran":0,"of":6,"unverified":6,"pointer_only":0}}],"papers_shown":4,"tasks":[{"task":"/task/self-supervised-image-classification","name":"Self-Supervised Image Classification","papers":2},{"task":"/task/semi-supervised-image-classification","name":"Semi-Supervised Image Classification","papers":2},{"task":"/task/adversarial-robustness","name":"Adversarial Robustness","papers":1},{"task":"/task/contrastive-learning","name":"Contrastive Learning","papers":1},{"task":"/task/fairness","name":"Fairness","papers":1},{"task":"/task/representation-learning","name":"Representation Learning","papers":1},{"task":"/task/self-supervised-learning","name":"Self-Supervised Learning","papers":1}],"tasks_shown":7,"n_tasks":7,"usage_by_year":[{"year":"2020","papers":1},{"year":"2021","papers":2},{"year":"2022","papers":1}],"row_source":"methods_table","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/simclrv2"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}