Graph Classification On Nci1

Metriken

Accuracy

Ergebnisse

Leistungsergebnisse verschiedener Modelle zu diesem Benchmark

		Paper Title
TFGW ADJ (L=2)	88.1%	Template based Graph Neural Network with Optimal Transport Distances
ESA (Edge set attention, no positional encodings)	87.835±0.644	An end-to-end attention-based approach for learning on graphs
WKPI-kmeans	87.2%	Learning metrics for persistence-based summaries and applications for graph classification
FGW wl h=4 sp	86.42%	Optimal Transport for structured data with application on graphs
WL-OA	86.1%	On Valid Optimal Assignment Kernels and Applications to Graph Classification
WL-OA Kernel	86.1%	Weisfeiler and Leman Go Neural: Higher-order Graph Neural Networks
FGW wl h=2 sp	85.82%	Optimal Transport for structured data with application on graphs
WWL	85.75%	Wasserstein Weisfeiler-Lehman Graph Kernels
DUGNN	85.50%	Learning Universal Adversarial Perturbations with Generative Models
δ-2-LWL	85.5%	Weisfeiler and Leman go sparse: Towards scalable higher-order graph embeddings
CIN++	85.3%	CIN++: Enhancing Topological Message Passing
CORE-WL	85.12%	Graph Kernels: A Survey
GraphGPS	85.110±1.423	Recipe for a General, Powerful, Scalable Graph Transformer
GAT	85.109±1.107	Graph Attention Networks
PIN	85.1%	Weisfeiler and Lehman Go Paths: Learning Topological Features via Path Complexes
PNA	84.964±1.391	Principal Neighbourhood Aggregation for Graph Nets
Norm-GN	84.87%	A New Perspective on the Effects of Spectrum in Graph Neural Networks
GIN	84.818±0.936	How Powerful are Graph Neural Networks?
CAN	84.5%	Cell Attention Networks
Propagation kernels (pk)	84.5%	Propagation kernels: efficient graph kernels from propagated information

0 of 67 row(s) selected.

Command Palette

Graph Classification On Nci1

Metriken

Ergebnisse