AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Mean Shift for Self-Supervised Learning

Soroush Abbasi Koohpayegani, Ajinkya Tejankar, Hamed Pirsiavash

ICCV 2021 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Data-efficient and robust self-supervision · 82 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use weak augmentation for the target view (w/s) instead of strong on both views (ablation)MSF · ImageNet linear evaltop-1 accuracy: 71.4 -> 72.4Improved
use weak augmentation for the target view (w/s) (ablation)BYOL-asym · ImageNet linear evaltop-1 accuracy: 69.3 -> 69.5No clear effect
use nearest neighbors in the target (MSF vs BYOL) with weak/weak augmentationMSF w/w · ImageNet linear evaltop-1 accuracy: 60.1 -> 66.3Improved
use 5 nearest neighbors as targets instead of only 1 (BYOL-asym) (ablation)MSF · ImageNet 1-NNtop-1 accuracy: 55.8 -> 62.0Improved
use 5 nearest neighbors as targets instead of only 1 (BYOL-asym) (ablation)MSF · ImageNet 20-NNtop-1 accuracy: 59.1 -> 64.9Improved
increase k from 5 to 10 (ablation)MSF w/s · ImageNet 20-NNtop-1 accuracy: 64.9 -> 65.7Improved
increase k from 5 to 50 (ablation)MSF w/s · ImageNet 1-NNtop-1 accuracy: 62.0 -> 61.5Hurt
reduce memory bank from 1M to 128K (ablation)MSF w/s · ImageNet 20-NNtop-1 accuracy: 64.9 -> 65.2No clear effect
reduce memory bank from 1M to 128K (ablation)MSF w/s · Linear transfer, 10 datasetsmean accuracy: 75.5 -> 76.3Improved
feed the same augmented view to both encoders (T1=T2) (ablation)MSF · ImageNettraining stabilityHurt
replace BYOL-asym by MSF (adding NN grouping), strong augmentationBYOL-asym · ImageNet linear evaltop-1 accuracy: 69.3 -> 71.4Improved
replace BYOL-asym by MSF (adding NN grouping)BYOL-asym · ImageNet 1% labelstop-1 accuracy: 47.9 -> 53.5Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy