Using two years of Lightning Network gossip data, a simple MLP predicts channel closure type (open/mutual/forced) with 0.38 macro F1, outperforming graph-based models.
Topological analysis of bitcoin’s lightning network
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Predicting Channel Closures in the Lightning Network with Machine Learning
Using two years of Lightning Network gossip data, a simple MLP predicts channel closure type (open/mutual/forced) with 0.38 macro F1, outperforming graph-based models.