Densities of almost surely terminating probabilistic programs are differentiable almost everywhere

DOI10.1007/978-3-030-72019-3_16zbMath1473.68048arXiv2004.03924OpenAlexW3148899495MaRDI QIDQ2233472

Yanyan Li

Publication date: 18 October 2021

Full work available at URL: https://arxiv.org/abs/2004.03924

Mathematics Subject Classification ID

Other programming paradigms (object-oriented, sequential, concurrent, automatic, etc.) (68N19) Mathematical aspects of software engineering (specification, verification, metrics, requirements, etc.) (68N30)

Related Items (3)

Correctness of sequential Monte Carlo inference for probabilistic programming languages ⋮ Densities of almost surely terminating probabilistic programs are differentiable almost everywhere ⋮ Bayesian strategies: probabilistic programs as generalised graphical models

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Semantics of probabilistic programs
• A type-theoretical alternative to ISWIM, CUCH, OWHY
• On the hardness of analyzing probabilistic programs
• The differential lambda-calculus
• Conservative set valued fields, automatic differentiation, stochastic gradient methods and deep learning
• Correctness of automatic differentiation via diffeologies and categorical gluing
• Densities of almost surely terminating probabilistic programs are differentiable almost everywhere
• Fine-grained semantics for probabilistic programs
• An introduction to manifolds. 2nd revised ed.
• Introduction to Smooth Manifolds
• A type theory for probability density functions
• A lambda-calculus foundation for universal probabilistic programming
• Deriving a probability density calculator (functional pearl)
• Contextual Equivalence for Probabilistic Programs with Continuous Random Variables and Scoring
• Commutative Semantics for Probabilistic Programming
• Probabilistic Programming Language and its Incremental Evaluation
• Symbolic execution and program testing
• Full Abstraction for Probabilistic PCF
• On the Versatility of Open Logical Relations
• Discontinuous Hamiltonian Monte Carlo for discrete parameters and discontinuous likelihoods
• An Introduction to Variational Autoencoders

This page was built for publication: Densities of almost surely terminating probabilistic programs are differentiable almost everywhere