Converging RGNNs express exactly the graded modal μ-calculus (μGML) and are equivalent to graded-bisimulation-invariant halting RGNNs; output-converging RGNNs express at least μGML.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
ACCEPT 1representative citing papers
citing papers explorer
-
On Halting vs Converging in Recurrent Graph Neural Networks
Converging RGNNs express exactly the graded modal μ-calculus (μGML) and are equivalent to graded-bisimulation-invariant halting RGNNs; output-converging RGNNs express at least μGML.