mlsanityData to 7 Oct 2026

Contrastive Learning Is Spectral Clustering On Similarity Graph

arXiv 2303.15103 ↗First posted 27 Mar 2023cs.LG cs.AI cs.CVcited by 24

Not on the map. Why?

Abstract (arXiv)

Contrastive learning is a powerful self-supervised learning method, but we have a limited theoretical understanding of how it works and why it works. In this paper, we prove that contrastive learning with the standard InfoNCE loss is equivalent to spectral clustering on the similarity graph. Using this equivalence as the building block, we extend our analysis to the CLIP model and rigorously characterize how similar multi-modal objects are embedded together. Motivated by our theoretical insights, we introduce the Kernel-InfoNCE loss, incorporating mixtures of kernel functions that outperform the standard Gaussian kernel on several vision datasets. The code is available at https://github.com/yifanzhang-pro/Kernel-InfoNCE.

Key facts. Cited by 24 papers in the map (data through 7 Oct 2026). Builds on 24 of its 42 references that are in the map. First posted on arXiv 27 Mar 2023.

Builds on

24 of 42 references.

References · 24

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  3. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  4. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  5. 2018Representation Learning with Contrastive Predictive Codingcited by 7,662
  6. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,482
  7. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 6,979
  8. 2020Improved Baselines with Momentum Contrastive Learningcited by 2,282
  9. 2020Understanding Contrastive Representation Learning through Alignment and Uniformity on the Hyperspherecited by 1,393
  10. 2020Graph Contrastive Learning with Augmentationscited by 1,203
  11. 2020What Makes for Good Views for Contrastive Learning?cited by 843
  12. 2020Contrastive Multi-View Representation Learning on Graphscited by 651
12 more
  1. 2019A Theoretical Analysis of Contrastive Unsupervised Representation Learningcited by 530
  2. 2020Hard Negative Mixing for Contrastive Learningcited by 399
  3. 2018Tune: A Research Platform for Distributed Model Selection and Trainingcited by 395
  4. 2021Provable Guarantees for Self-Supervised Deep Learning with Spectral Contrastive Losscited by 311
  5. 2021Parametric Contrastive Learningcited by 217
  6. 2022Contrastive and Non-Contrastive Self-Supervised Learning Recover Global and Local Spectral Embedding Methodscited by 146
  7. 2022Chaos is a Ladder: A New Theoretical Understanding of Contrastive Learning via Augmentation Overlapcited by 105
  8. 2020Deep Neural Tangent Kernel and Laplace Kernel Have the Same RKHScited by 74
  9. 2022Understanding Deep Contrastive Learning via Coordinate-wise Optimizationcited by 35
  10. 2022Your Contrastive Learning Is Secretly Doing Stochastic Neighbor Embeddingcited by 28
  11. 2022Neural Eigenfunctions Are Structured Representation Learnerscited by 16
  12. 2022A Probabilistic Graph Coupling View of Dimension Reductioncited by 13

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 5; 2024: 7; 2025: 9; 2026: 3.

24 papers.

Sort:

By area:

  1. 2023Matrix Information Theory for Self-Supervised Learningcited by 21
  2. 2024Non-negative Contrastive Learningcited by 18
  3. 2023Information Flow in Self-Supervised Learningcited by 16
  4. 2024Provable Contrastive Continual Learningcited by 13
  5. 2024Revisiting Self-Supervised Heterogeneous Graph Learning from Spectral Clustering Perspectivecited by 7
  6. 2023OTMatch: Improving Semi-Supervised Learning with Optimal Transportcited by 6
  7. 2024Tight PAC-Bayesian Risk Certificates for Contrastive Learningcited by 6
  8. 2024THESAURUS: Contrastive Graph Clustering by Swapping Fused Gromov-Wasserstein Couplingscited by 5
  9. 2025Existing LLMs Are Not Self-Consistent For Simple Taskscited by 5
  10. 2025LeakyCLIP: Extracting Training Data from CLIPcited by 4
  11. 2025PCA++: How Uniformity Induces Robustness to Background Noise in Contrastive Learningcited by 4
  12. 2024Lambda: Learning Matchable Prior For Entity Alignment with Unlabeled Dangling Casescited by 3
12 more
  1. 2023COSTAR: Improved Temporal Counterfactual Estimation with Self-Supervised Learningcited by 2
  2. 2025Difficult Examples Hurt Unsupervised Contrastive Learning: A Theoretical Perspectivecited by 2
  3. 2025A Markov Categorical Framework for Language Modelingcited by 2
  4. 2023Deep Tensor Networkcited by 0
  5. 2024Enhancing Graph Self-Supervised Learning with Graph Interplaycited by 0
  6. 2025Affect-aware Cross-Domain Recommendation for Art Therapy via Music Preference Elicitationcited by 0
  7. 2025Enhancing Contrastive Link Prediction With Edge Balancing Augmentationcited by 0
  8. 2025Understanding InfoNCE: Transition Probability Matrix Induced Feature Clusteringcited by 0
  9. 2025Graph Contrastive Learning via Spectral Graph Alignmentcited by 0
  10. 2026Maximizing Incremental Information Entropy for Contrastive Learningcited by 0
  11. 2026Expressivity In Multimodal Contrastive Learningcited by 0
  12. 2026Refining Heuristic-Based Bitcoin Address Clustering with Graph Neural Networkscited by 0