Normalizing-flow kernels generate coarse-grained forces from configuration-only data, reducing local distortions found with Gaussian noise kernels while preserving global conformational accuracy.
At each layer the input x = (x1,x2) is partitioned into two blocks, and one block is transformed conditioned on the other: y1 = x1, y2 = li θ (x2|x1)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
physics.chem-ph 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Operator Forces For Coarse-Grained Molecular Dynamics
Normalizing-flow kernels generate coarse-grained forces from configuration-only data, reducing local distortions found with Gaussian noise kernels while preserving global conformational accuracy.