LResNet performs hyperbolic residual connections with a normalized weighted sum (the Lorentzian centroid), avoiding tangent-space mappings and improving efficiency and accuracy in hyperbolic GNNs, graph transformers, and CNNs.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Lorentzian Residual Neural Networks
LResNet performs hyperbolic residual connections with a normalized weighted sum (the Lorentzian centroid), avoiding tangent-space mappings and improving efficiency and accuracy in hyperbolic GNNs, graph transformers, and CNNs.