AI Paper Atlas

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

Self-Supervised Learning with Kernel Dependence Maximization

Yazhe Li, Roman Pogodin, Danica J. Sutherland, Arthur Gretton

NeurIPS 2021 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Data-efficient and robust self-supervision · 21 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 a target network to SSL-HSIC (ablation)SSL-HSIC · ImageNet val linear evaluationTop-1 accuracy: 72.2 -> 74.8Improved
Add a target network to SSL-HSIC (ablation)SSL-HSIC · ImageNet val fine-tuning (1% labels)Top-1 accuracy: 45.3 -> 52.1Improved
Use SSL-HSIC regularization instead of InfoNCE regularization (ablation)SSL-HSIC · ImageNet linear evaluation, 100 epochsTop-1 accuracy: 66.0 -> 66.7Improved
Use a nonlinear latent kernel rather than a linear kernel (ablation)SSL-HSIC · ImageNet linear evaluation, 100 epochsTop-1 accuracy: 65.27 -> 66.67Improved
Use an IMQ kernel instead of a Gaussian kernel (ablation)SSL-HSIC · ImageNet linear evaluation, 100 epochsTop-1 accuracy: 66.67 -> 66.72No clear effect
Increase the number of random Fourier features (ablation)SSL-HSIC · ImageNet linear evaluation, 100 epochsTop-1 accuracy: 66.0 -> 66.4Improved
Use 512 random Fourier features rather than the exact kernel (ablation)SSL-HSIC · ImageNet linear evaluation, 100 epochsTop-1 accuracy: 66.7 -> 66.4Hurt
Increase training batch size from 256 to 1024 (ablation)SSL-HSIC · ImageNet linear evaluation, 100 epochsTop-1 accuracy: 63.7 -> 66.7Improved
Increase batch size from 1024 to 4096 (ablation)SSL-HSIC · ImageNet linear evaluation, 100 epochsTop-1 accuracy: 66.7 -> 66.7No clear effect
Increase projector/predictor output dimension from 64 to 256 (ablation)SSL-HSIC · ImageNet linear evaluation, 100 epochsTop-1 accuracy: 65.4 -> 66.4Improved
Increase backbone capacity from ResNet-50 (1x) to ResNet-50 (4x) (ablation)SSL-HSIC · ImageNet linear evaluationTop-1 accuracy: 74.8 -> 79.1Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 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