Fast Rates for Semi-Supervised Learning via Data-Augmentation Graph Regularization
Adam M. Oberman
cs.LG, math.ST, stat.ML
Submitted: 2026-07-08
Comments: 23 pages, 2 figures
Code: https://github.com/adam-oberman/fast-rates-ssl
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- A Theoretical Analysis of Contrastive Unsupervised Representation Learning
- Contrastive and Non-Contrastive Self-Supervised Learning Recover Global and Local Spectral Embedding Methods
- Sharper bounds for uniformly stable algorithms
- How does Labeling Error Impact Contrastive Learning? A Perspective from Data Dimensionality Reduction
- A Simple Framework for Contrastive Learning of Visual Representations
- Big Self-Supervised Models are Strong Semi-Supervised Learners
- Stability Analysis and Learning Bounds for Transductive Regression Algorithms
- Cluster-aware Semi-supervised Learning: Relational Knowledge Distillation Provably Learns Clustering
- Generalization Bounds for Uniformly Stable Algorithms
- High probability generalization bounds for uniformly stable algorithms with nearly optimal rate
- Provable Guarantees for Self-Supervised Deep Learning with Spectral Contrastive Loss
- Data-Efficient Image Recognition with Contrastive Predictive Coding
- Generalization Analysis for Contrastive Representation Learning
- Label Propagation with Weak Supervision
- Information-Theoretic Generalization Bounds for Transductive Learning and its Applications
- Can semi-supervised learning use all the data effectively? A lower bound perspective
- Stability and Generalization of Graph Convolutional Neural Networks
- Chaos is a Ladder: A New Theoretical Understanding of Contrastive Learning via Augmentation Overlap
- Sample Efficiency of Data Augmentation Consistency Regularization
- Understanding Augmentation-based Self-Supervised Representation Learning via RKHS Approximation and Regression
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks