Papers › VICReg: Variance-Invariance-Covariance Regularization for Self-Supervised Learning

VICReg: Variance-Invariance-Covariance Regularization for Self-Supervised Learning

11 May 2021NeurIPS 2021 12arXiv:2105.04906archive 2025-07-28

Adrien Bardes, Jean Ponce, Yann Lecun

Recent self-supervised methods for image representation learning are based on maximizing the agreement between embedding vectors from different views of the same image. A trivial solution is obtained when the encoder outputs constant vectors. This collapse problem is often avoided through implicit biases in the learning architecture, that often lack a clear justification or interpretation. In this paper, we introduce VICReg (Variance-Invariance-Covariance Regularization), a method that explicitly avoids the collapse problem with a simple regularization term on the variance of the embeddings along each dimension individually. VICReg combines the variance term with a decorrelation mechanism based on redundancy reduction and covariance regularization, and achieves results on par with the state of the art on several downstream tasks. In addition, we show that incorporating our new variance term into other methods helps stabilize the training and leads to performance improvements.

PaperPDFConference PDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2105.04906")

Code

Syntology Ran 11 of 17 code samples harvested from 4 repositories linked to this paper; 6 have no recorded run. Of those that ran: 1 ran · honoured contract; 1 ran · violated contract; 3 ran · our draft was wrong; 2 ran · fixture could not drive it; 4 ran with no contract checked.

By repository: official repository: 5 samples from 1 repository, 4 ran; community (archive-listed): 11 samples from 3 repositories, 6 ran; 1 identical to code first harvested elsewhere. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

facebookresearch/vicreg officialmentioned on GitHubpytorch report
AnnaManasyan/VICReg mentioned on GitHubpytorchMIT report
FloCF/SSL_pytorch mentioned on GitHubpytorch report
lightly-ai/lightly mentioned on GitHubpytorch report
vturrisi/solo-learn mentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

17 samples harvested; 11 ran; 1 honoured the contract we drafted; 6 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

1ran · honoured contract
1ran · violated contract
3ran · our draft was wrong
2ran · fixture could not drive it
4ran
6unverified

Licence: 1 of the 17 samples is pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 4 repositories linked to this paper, official or community; each sample names its own and says which. Some samples are identical code Syntology first harvested from another repository; for those, this paper's copy is not located and its licence is not recorded. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

FullGatherLayer facebookresearch/vicreg/main_vicreg.py official repository ran MIT (permissive) · 90bc3c3bb8e6523b · report
Projector facebookresearch/vicreg/main_vicreg.py official repository ran · our draft was wrong MIT (permissive) · 19322d1e968e9137 · report
adjust_learning_rate facebookresearch/vicreg/main_vicreg.py official repository ran · honoured contract MIT (permissive) · 0a59c9ccf5135868 · report
exclude_bias_and_norm facebookresearch/vicreg/main_vicreg.py official repository ran · violated contract fingerprinted MIT (permissive) · 28c640db8f721c0d · report
VICReg facebookresearch/vicreg/main_vicreg.py official repository unverified MIT (permissive) · a8066cb8d5d3e19d · report
GatherLayer lightly-ai/lightly/lightly/loss/vicreg_loss.py community (archive-listed) ran MIT (permissive) · 4073146360f534c5 · report
VICReg FloCF/SSL_pytorch/torchselfsup/models/vicreg.py community (archive-listed) ran · metamorphic tier: invariant MIT (permissive) · c796daf86600fb9e · report
VICRegLoss lightly-ai/lightly/lightly/loss/vicreg_loss.py community (archive-listed) ran fingerprinted MIT (permissive) · 23452a5a035b134a · report
invariance_loss lightly-ai/lightly/lightly/loss/vicreg_loss.py community (archive-listed) ran · our draft was wrong fingerprinted MIT (permissive) · 48e54f52f61ab9b8 · report
off_diagonal AnnaManasyan/VICReg/loss.py community (archive-listed) ran · fixture could not drive it fingerprinted MIT (permissive) · 3e30d88eaef01190 · report
variance_loss lightly-ai/lightly/lightly/loss/vicreg_loss.py community (archive-listed) ran · our draft was wrong fingerprinted MIT (permissive) · adb0bbb17650f6fe · report
augment AnnaManasyan/VICReg/utils.py community (archive-listed) unverified MIT (permissive) · 73fbb91c4b92324b · report
cov_loss AnnaManasyan/VICReg/loss.py community (archive-listed) unverified MIT (permissive) · 0c5a6989161b85e9 · report
gather lightly-ai/lightly/lightly/loss/vicreg_loss.py community (archive-listed) unverified MIT (permissive) · f4b990248d9c5170 · report
optim AnnaManasyan/VICReg/utils.py community (archive-listed) unverified MIT (permissive) · 1eeada3f42590ae7 · report
std_loss AnnaManasyan/VICReg/loss.py community (archive-listed) unverified MIT (permissive) · 62a1a07b187aa7a9 · report
accuracy identical code first harvested elsewhere ran · fixture could not drive it licence of this copy not recorded · b0f936d4d6ae3b8c · report

Tasks

Representation LearningSelf-Supervised Image ClassificationSelf-Supervised LearningSemi-Supervised Image Classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Self-Supervised Image Classification ImageNet VICReg (ResNet50) Number of Params 24M #90 of 144 Archive leaderboard report
Self-Supervised Image Classification ImageNet VICReg (ResNet50) Top 1 Accuracy 73.2 #90 of 144 Archive leaderboard report
Self-Supervised Image Classification ImageNet VICReg (ResNet50) Top 5 Accuracy 91.1 #90 of 144 Archive leaderboard report
Semi-Supervised Image Classification ImageNet - 1% labeled data VICREG (Resnet-50) Top 1 Accuracy 54.8% #47 of 65 Archive leaderboard report
Semi-Supervised Image Classification ImageNet - 1% labeled data VICREG (Resnet-50) Top 5 Accuracy 79.4% #47 of 65 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections