FuncGNN, a GNN with hybrid aggregation, ratio-conditioned normalization, and dense layer fusion, reports state-of-the-art MAE on signal probability and truth-table distance prediction for AIG circuits.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
FuncGNN: Learning Functional Semantics of Logic Circuits with Graph Neural Networks
FuncGNN, a GNN with hybrid aggregation, ratio-conditioned normalization, and dense layer fusion, reports state-of-the-art MAE on signal probability and truth-table distance prediction for AIG circuits.