Research map / Label-efficient and robust learning / Self-supervised and contrastive
Toward Understanding the Feature Learning Process of Self-supervised Contrastive Learning
Zixin Wen, Yuanzhi Li
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use RandomMask to create disjoint positive views | Contrastive learning with a one-hidden-layer ReLU network · Sparse coding model, downstream classification | accuracy: o(1) -> 1 - o(1) | Improved |
| Remove data augmentation (ablation) | Contrastive learning with a one-hidden-layer ReLU network · Sparse coding model, downstream classification | accuracy: 1 - o(1) -> o(1) | Hurt |
| Use crop-resize without color distortion (ablation) | Contrastive learning on CIFAR10 · CIFAR10 representations | feature sparsity | Hurt |
| Use standard augmentations including color distortion | Contrastive learning on CIFAR10 · CIFAR10 representations | feature sparsity | Improved |
Closest papers by meaning
- The Power of Contrast for Feature Learning: A Theoretical Analysis (JMLR 2023 · similarity 0.89)
- Which Features are Learnt by Contrastive Learning? On the Role of Simplicity Bias in Class Collapse and Feature Suppression (ICML 2023 · similarity 0.88)
- An Augmentation-Aware Theory for Self-Supervised Contrastive Learning (ICML 2025 · similarity 0.88)
- Towards the Generalization of Contrastive Self-Supervised Learning (ICLR 2023 · similarity 0.88)
- ArCL: Enhancing Contrastive Learning with Augmentation-Robust Representations (ICLR 2023 · similarity 0.88)
- Self-Supervised Contrastive Learning is Approximately Supervised Contrastive Learning (NeurIPS 2025 · similarity 0.86)
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.
