Pith. sign in

REVIEW 6 cited by

Deep Gaussian Embedding of Graphs: Unsupervised Inductive Learning via Ranking

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1707.03815 v4 pith:KUF55U34 submitted 2017-07-12 stat.ML cs.LGcs.SI

classification stat.MLcs.LGcs.SI
keywords nodenetworknodesgraphslearnlearningtasksapproach
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Methods that learn representations of nodes in a graph play a critical role in network analysis since they enable many downstream learning tasks. We propose Graph2Gauss - an approach that can efficiently learn versatile node embeddings on large scale (attributed) graphs that show strong performance on tasks such as link prediction and node classification. Unlike most approaches that represent nodes as point vectors in a low-dimensional continuous space, we embed each node as a Gaussian distribution, allowing us to capture uncertainty about the representation. Furthermore, we propose an unsupervised method that handles inductive learning scenarios and is applicable to different types of graphs: plain/attributed, directed/undirected. By leveraging both the network structure and the associated node attributes, we are able to generalize to unseen nodes without additional training. To learn the embeddings we adopt a personalized ranking formulation w.r.t. the node distances that exploits the natural ordering of the nodes imposed by the network structure. Experiments on real world networks demonstrate the high performance of our approach, outperforming state-of-the-art network embedding methods on several different tasks. Additionally, we demonstrate the benefits of modeling uncertainty - by analyzing it we can estimate neighborhood diversity and detect the intrinsic latent dimensionality of a graph.

Discussion (0). Sign in to comment.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Delving into Instance-Dependent Label Noise in Graph Data: A Comprehensive Study and Benchmark

    cs.LG 2025-06 conditional novelty 7.0 of 10

    BeGIN provides 10 graph datasets with six label-noise types and a broad evaluation, showing that LLM-simulated noise hurts GNNs more than uniform or pairwise noise.

  2. WATS: Calibrating Graph Neural Networks with Wavelet-Aware Temperature Scaling

    cs.LG 2025-06 conditional novelty 5.0 of 10

    WATS uses heat-kernel graph wavelet features to predict node-specific temperatures for GNN calibration, reporting the lowest ECE on benchmark datasets, though hyperparameter selection uses test-set information.

  3. DistShap: Scalable GNN Explanations with Distributed Shapley Values

    cs.LG 2025-06 conditional novelty 5.0 of 10

    DistShap parallelizes Shapley-based GNN explanations across many GPUs, scaling to graphs with tens of millions of edges while preserving explanation quality.

  4. Sphere of Influence Centrality via Shapley Values: Empirical Approximation and Network Coverage Analysis

    cs.SI 2026-06 unverdicted novelty 4.0 of 10

    Empirical tests on three real networks show Shapley-value node selection for coverage under reachability rules reaches ~0.9 approximation ratio and beats degree baseline, with one case covering half of Cora using 26 nodes.

  5. NGAT: A Node-level Graph Attention Network for Long-term Stock Prediction

    q-fin.ST 2025-07 conditional novelty 4.0 of 10

    NGAT, a graph attention network with node-specific attention parameters, improves long-term return and volatility forecasts over LSTM and GAT baselines on the SPNews and ACL2018 datasets.

  6. Directed Link Prediction using GNN with Local and Global Feature Fusion

    cs.LG 2025-06 reject novelty 3.0 of 10

    A GNN that fuses path labels, community labels, and contrastive embeddings on a directed line graph is reported to beat six baselines on directed link prediction, but its proof has errors.

Pith tools