Normalizing flows are constructed by learning the velocity of a stochastic interpolant via a quadratic loss derived from its probability current, yielding an efficient ODE-based alternative to diffusion models.
Action matching: A variational method for learning stochastic dynamics from samples
1 Pith paper cite this work, alongside 1 external citations. Polarity classification is still indexing.
1
Pith paper citing it
1
external citations · external index
fields
cs.LG 1years
2022 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Building Normalizing Flows with Stochastic Interpolants
Normalizing flows are constructed by learning the velocity of a stochastic interpolant via a quadratic loss derived from its probability current, yielding an efficient ODE-based alternative to diffusion models.