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

SSOLE: Rethinking Orthogonal Low-rank Embedding for Self-Supervised Learning

ICLR 2025 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Data-efficient and robust self-supervision

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.

Revisits orthogonal low-rank embedding (OLE) and adapts it into SSOLE, a self-supervised learning objective for multi-view representation learning.

Problem

Directly applying Orthogonal Low-rank Embedding (OLE) to self-supervised learning faces challenges, as the abstract is truncated after noting the first issue (the virtual ... ), so the specific obstacles are not fully stated.

Approach

Adapts OLE, which enhances intra-class similarity in a low-rank subspace and inter-class dissimilarity in a high-rank subspace, to unlabeled multi-view SSL, addressing the challenges of applying it directly.

Experimental setup

Builds on: Orthogonal Low-rank Embedding

Closest papers by meaning

• Views Can Be Deceiving: Improved SSL Through Feature Space Augmentation (ICLR 2024 · similarity 0.86)
• Joint‑Embedding vs Reconstruction: Provable Benefits of Latent Space Prediction for Self‑Supervised Learning (NeurIPS 2025 · similarity 0.86)
• Non-parametric Representation Learning with Kernels (AAAI 2024 · similarity 0.86)
• Can Generative Models Improve Self-Supervised Representation Learning? (AAAI 2025 · similarity 0.86)
• Clustering Properties of Self-Supervised Learning (ICML 2025 · similarity 0.86)
• InterLUDE: Interactions between Labeled and Unlabeled Data to Enhance Semi-Supervised Learning (ICML 2024 · similarity 0.85)