Autoencoder pretraining of graph mesh encoders makes neural radar simulators less sensitive to shape-preserving changes in mesh topology.
Deep stochastic radar models,
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
-
Reducing the Sensitivity of Neural Physics Simulators to Mesh Topology via Pretraining
Autoencoder pretraining of graph mesh encoders makes neural radar simulators less sensitive to shape-preserving changes in mesh topology.