AI Paper Atlas

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

On the Generalization and Causal Explanation in Self-Supervised Learning

Wenwen Qiang, Zeen Song, Ziyin Gu, Jiangmeng Li, Changwen Zheng, Fuchun Sun, Hui Xiong

IJCV 2024 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Data-efficient and robust self-supervision · 4 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
add UMM fine-tuningSimSiam (ResNet-18) · CIFAR-10 linearaccuracy: 91.71 -> 92.81Improved
add UMM fine-tuningBYOL (ResNet-18) · STL-10 linearaccuracy: 91.99 -> 93.99Improved
add UMM fine-tuningVICRegL (ResNet-18) · CIFAR-100 linearaccuracy: 68.03 -> 69.21Improved
add UMM fine-tuningBYOL (ResNet-50) · ImageNet-100 linear top-1accuracy: 75.66 -> 76.98Improved
add UMM fine-tuningRELIC v2 (ResNet-50) · ImageNet linear, 1000 epochstop-1 accuracy: 77.1 -> 78.2Improved
add UMM fine-tuningSimCLR (ResNet-50) · ImageNet 1% labels semi-supervised, 1000 epochstop-1 accuracy: 48.3 -> 52.6Improved
add UMM fine-tuningMoCo (C4) · COCO detectionAP: 39.3 -> 40.7Improved
add coding-rate-reduction term to SSL plus JS alignment term (full UMM) (ablation)SimSiam (ResNet-18) · CIFAR-10 linearaccuracy: 91.91 -> 92.81Improved
add JS alignment term to SSL plus coding-rate-reduction term (full UMM) (ablation)SimSiam (ResNet-18) · CIFAR-10 linearaccuracy: 91.97 -> 92.81Improved
add coding-rate-reduction term to SSL plus JS term (full UMM) (ablation)SimCLR (ResNet-18) · CIFAR-10 linearaccuracy: 92.03 -> 92.93Improved
add only JS term to SSL loss (ablation)Barlow Twins (ResNet-18) · CIFAR-10 linearaccuracy: 90.88 -> 90.91No clear effect
vary coding-rate epsilon from 0.1 to 0.9 (ablation)SimCLR + UMM · CIFAR-10 linearaccuracy: 92.93 -> 92.92No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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