Graph-Based Semi-Supervised Learning

Also known as SSL

The field studies how a graph connecting labeled and unlabeled examples can reveal structure that helps classify nodes or data points when labels are scarce. Methods include harmonic-function and label-propagation algorithms, Laplacian regularization, graph neural networks, and pseudo-labeling, sometimes with active querying or learned graph construction. These approaches are used in settings such as citation-network node classification, image and molecular classification, and streaming data, with work also targeting faster solvers and better calibration. A central challenge is that gains depend on graph quality: noisy edges, changing distributions, or weak agreement between neighbors can mislead propagation, while stronger smoothing can erase useful distinctions.

Latest papers

33 papers4 papers in the last four weeks, up 33% on the four weeks before. 0.0% of all new papers.

Fractional Heat Kernel for Semi-Supervised Graph Learning with Small Training Sample Size
Oct 6, 2025Farid Bozorgnia, Vyacheslav Kungurtsev, Shirali Kadyrov +1Over-Smoothing in GNNs Graph-Based Semi-Supervised Learning
GLL: A Differentiable Graph Learning Layer for Neural Networks
Dec 11, 2024Jason Brown, Bohan Chen, Harris Hardiman-Mostow +2Graph-Based Label Propagation Graph-Based Semi-Supervised Learning
Department of Mathematics University of California, Los Angeles Los Angeles, CA 90095, USA · Computing + Mathematical Sciences (CMS) Department California Institute of Technology Pasadena, CA 91125, USA · School of Mathematics University of Minnesota Minneapolis, MN 55455, USA
STAMP: Symmetric Temporal Averaging Message Passing for Domain Adaptation in Chronologically Split Graphs
Nov 17, 2024Joo Young Park, Sejun Park, Hyunwoo ParkTemporal Domain Adaptation Domain Adaptation
College of Medicine, Seoul National University, Gwanak-gu, Seoul 08826, Korea · Electrical and Computer Engineering, Seoul National University, Gwanak-gu, Seoul 08826, Korea · Graduate School of Data Science, Seoul National University, Gwanak-gu, Seoul 08826, Korea

Oct 6, 2025J/K move · Enter open · S save

Fractional Heat Kernel for Semi-Supervised Graph Learning with Small Training Sample Size

Farid Bozorgnia, Vyacheslav Kungurtsev, Shirali Kadyrov, Mohsen Yousefnezhad

Over-Smoothing in GNNs Graph-Based Semi-Supervised Learning Graph-Based Label Propagation Graph Diffusion Graph Neural Networks

Open paper AskPDF arXiv

We develop a source-driven fractional heat-kernel framework for semi-super-vised graph learning that combines nonlocal propagation with sustained label information. A fixed nonzero label source compatible with the Laplacian null space prevents asymptotic collapse into that space, providing a mechanism for mitigating oversmoothing at long diffusion times. The fractional order controls the relative modal attenuation and the spectral weighting of the sustained response, while the diffusion time sets the propagation horizon. We characterize conservation laws and equilibria on normalized and disconnected graphs, develop a null-space deflation, and analyze the approximation of the propagators. On Two-Moon, fractional orders improve source-free propagation, while compatible source-driven diffusion exceeds

96\%

mean accuracy with one training label per class at orders

0.8

and

1

. On Cora and CiteSeer, the source-driven pipeline improves mean accuracy over GAT by

9.2

and

8.0

percentage points at one training label per class, with model selection on

500

labeled validation nodes and closely comparable classical and fractional pipeline configurations; it matches GAT on PubMed and trails it at ten and twenty labels per class on Cora. Within GraphHeat, validation-based exponent selection at a common diffusion time yields a paired gain of

1.45

percentage points at one label per class.

Latest papers

Fractional Heat Kernel for Semi-Supervised Graph Learning with Small Training Sample Size

GLL: A Differentiable Graph Learning Layer for Neural Networks

STAMP: Symmetric Temporal Averaging Message Passing for Domain Adaptation in Chronologically Split Graphs