Universal Function Approximation on Graphs

Authors: Rickard Brüel Gabrielsson

NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In this work we produce a framework for constructing universal function approximators on graph isomorphism classes. We prove how this framework comes with a collection of theoretically desirable properties and enables novel analysis. We show how this allows us to achieve state-of-the-art performance on four different well-known datasets in graph classification and separate classes of graphs that other graph-learning methods cannot.
Researcher Affiliation Academia Rickard Brüel-Gabrielsson rbg@cs.stanford.edu
Pseudocode Yes Algorithm 1 Subset Parsing Algorithm; Algorithm 2 Node Parsing Algorithm (NPA); Algorithm 3 Node Parsing Baseline Algorithm (NPBA)
Open Source Code Yes The complexity of the underlying algorithm is O(#edges #nodes) and code is publicly available1. 1https://github.com/bruel-gabrielsson/universal-function-approximation-on-graphs
Open Datasets Yes See Table 1 for results on graph classification benchmarks. ... Datasets: NCI1 MUTAG PROTEINS PTC
Dataset Splits Yes We report average and standard deviation of validation accuracies across the 10 folds within the cross-validation.
Hardware Specification No The paper does not specify any hardware details such as GPU/CPU models, memory, or specific computing environments used for the experiments.
Software Dependencies No The paper does not provide specific version numbers for any software dependencies. It mentions using "Neural Networks" and "LSTM" but without version details.
Experiment Setup Yes In the experiments, the W(G) features are summed and passed to a classifier consisting of fully connected NNs. For NPA, sv sorts randomly, but with "-S", sv sorts based on the levels of subgraphs S1 and S2. For subgraph dropout "-D" we use K = 5.