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

AUC-CL: A Batchsize-Robust Framework for Self-Supervised Contrastive Representation Learning

ICLR 2024 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Contrastive versus non-contrastive theory

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes AUC-CL, a batch-size-robust contrastive self-supervised learning framework; only the abstract was available, so details are limited.

Problem

Contrastive self-supervised representation learning methods are sensitive to batch size, limiting their performance when batch sizes are small.

Approach

Framework framed around AUC (area under the curve) to make contrastive learning robust to batch size; the abstract alone does not give the mechanism in detail.

Limitations

• Full text unavailable; analysis limited to abstract

Closest papers by meaning

• Supervised Contrastive Learning (NeurIPS 2020 · similarity 0.92)
• Why do We Need Large Batchsizes in Contrastive Learning? A Gradient-Bias Perspective (NeurIPS 2022 · similarity 0.91)
• A Simple Framework for Contrastive Learning of Visual Representations (ICML 2020 · similarity 0.91)
• Self-Supervised Contrastive Learning is Approximately Supervised Contrastive Learning (NeurIPS 2025 · similarity 0.89)
• BECLR: Batch Enhanced Contrastive Few-Shot Learning (ICLR 2024 · similarity 0.88)
• Variational Supervised Contrastive Learning (NeurIPS 2025 · similarity 0.87)