Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Computationally-efficient Graph Modeling with Refined Graph Random Features

About

We propose refined GRFs (GRFs++), a new class of Graph Random Features (GRFs) for efficient and accurate computations involving kernels defined on the nodes of a graph. GRFs++ resolve some of the long-standing limitations of regular GRFs, including difficulty modeling relationships between more distant nodes. They reduce dependence on sampling long graph random walks via a novel walk-stitching technique, concatenating several shorter walks without breaking unbiasedness. By applying these techniques, GRFs++ inherit the approximation quality provided by longer walks but with greater efficiency, trading sequential, inefficient sampling of a long walk for parallel computation of short walks and matrix-matrix multiplication. Furthermore, GRFs++ extend the simplistic GRFs walk termination mechanism (Bernoulli schemes with fixed halting probabilities) to a broader class of strategies, applying general distributions on the walks' lengths. This improves the approximation accuracy of graph kernels, without incurring extra computational cost. We provide empirical evaluations to showcase all our claims and complement our results with theoretical analysis.

Krzysztof Choromanski, Avinava Dubey, Arijit Sehanobish, Isaac Reid• 2025

Related benchmarks

TaskDatasetResultRank
Graph ClassificationPROTEINS
Accuracy74.87
1383
Graph ClassificationMUTAG
Accuracy87.8
1229
Graph ClassificationNCI1
Accuracy75.47
707
Graph ClassificationCOLLAB
Accuracy73.6
532
Graph ClassificationENZYMES
Accuracy40.9
419
Graph ClassificationDD
Accuracy73.67
309
Graph ClassificationPTC-MR
Accuracy61.87
271
Graph ClassificationD&D
Accuracy74.5
179
Graph ClassificationIMDB MULTI
Accuracy49.8
168
Image ClassificationImageNet (val)
Top-1 Accuracy80.31
165
Showing 10 of 29 rows

Other info

Follow for update