Gradient estimation of probabilistic programs reduces soundly to probabilistic inference after programmable coupling and factorization, enabling new low-variance estimators that beat baselines.
Pearlmutter, Alexey Andreyevich Radul, and Jeffrey Mark Siskind
1 Pith paper cite this work, alongside 84 external citations. Polarity classification is still indexing.
1
Pith paper citing it
84
external citations · external index
fields
cs.PL 1years
2026 1verdicts
ACCEPT 1representative citing papers
citing papers explorer
-
GradInf: Gradient Estimation as Probabilistic Inference
Gradient estimation of probabilistic programs reduces soundly to probabilistic inference after programmable coupling and factorization, enabling new low-variance estimators that beat baselines.