LAFI 2025
who it's for
probabilistic programming researchers, programming language researchers, machine learning researchers, compiler engineers, runtime and systems engineers, Bayesian statisticians
topics
nested discrete probabilistic inference; probabilistic pushdown automata; optimistic value iteration bounds; atomic sheaf sampling semantics; memo PPL with JAX compilation; state space models in Turing.jl; data-oriented Rust backend for PPLs; data-parallel differentiation by optic composition; NP-NUTS nonparametric No-U-Turn sampler; neural surrogates for experimental design; lazy knowledge compilation for discrete PPLs; runtime adaptable Java probabilistic language
| title | author | length |
|---|---|---|
| Exact Inference for Nested Discrete Probabilistic Programs | Francesco Pontiggia, Ezio Bartocci, Michele Chiari | 09m |
Reduces nested discrete probabilistic programs to a probabilistic pushdown automata and then to a monotone system of polynomial equations whose variables are… | ||
| Reasoning About Sampling Without Sampling: Atomic Machines for Contextual Equivalence(…) | Anthony D'Arienzo, Jon M. Aytac | 13m |
Claims an abstract-machine semantics for probabilistic programs that treats sampling as deterministic up to atomic equivalence, using atomic sheaf / forcing… | ||
| A Domain-Specific PPL for Reasoning about Reasoning (or: a memo on memo) | Kartik Chandra, Tony Chen, Joshua B. Tenenbaum, Jonathan Ragan-Kelley | 20m |
Introduces memo, a Python-embedded domain-specific PPL that replaces sample/infer with choose/observe semantics and compiles recursive theory-of-mind models… | ||
| State Space Model Programming in Turing.jl | Tim Hargreaves, Qing Li, Charles Knipp, Frederic Wantiez, Simon J. Godsill, Hong Ge | 18m |
Two Julia packages, SSMProblems.jl and GeneralisedFilters.jl, provide a composable SSM API inside the Turing.jl ecosystem that implements Kalman, particle and… | ||
| Data-oriented Design for Differentiable, Probabilistic Programming | Owen Lynch, Maria-Nicoleta Crăciun, Younesse Kaddar, Sam Staton | 20m |
Reimplementation of the lazy rose tree from LazyPPL into a strict, data-oriented Rust backend reduces thunk size from about 16 to 4 bytes, stores nodes and… | ||
| Data-Parallel Differentiation by Optic Composition | Paul Wilson, Fabio Zanasi | 17m |
Provides a concrete open hypergraphs representation (struct-of-arrays with segmented arrays) and parallel algorithms for tensor and composition via… | ||
| Programming Languages as Core Technology for AI | Jack Feser | 08m |
Basis describes a concrete stack: Cairo causal PPL built on Pyro, an effectful Python library for algebraic effects, and program synthesis integrated into the… | ||
| NP-NUTS: A Nonparametric No-U-Turn Sampler | Maria-Nicoleta Crăciun, C.-H. Luke Ong, Sam Staton, Matthijs I. L. Vákár | 16m |
A new algorithm, NP‑NUTS, automatically adapts HMC trajectory length for models whose parameter dimension varies by combining NUTS’s doubling/tree construction… | ||
| Semantics of the memo Probabilistic Programming Language | Kartik Chandra, Nada Amin, Yizhou Zhang | 17m |
Formal semantics for the memo probabilistic programming language claim: a frame of mind is the core state that records choices, subframes and equations and… | ||
| Modern Bayesian Experimental Design | Desi R. Ivavona | 39m |
Presents concrete methods: learn neural surrogates to estimate the intractable expected information gain (using nested Monte Carlo and variational upper/lower… | ||
| Sandwood: Runtime Adaptable Probabilistic Programming for Java | Daniel Goodman, Adam Pocock, Natalia Kosilova | 19m |
Sandwood is an open‑source Java‑like probabilistic programming language, compiler and runtime that compiles models to Java classes whose instances can switch… | ||
| Lazy Knowledge Compilation for Discrete PPLs | Maddy Bowers, Alexander K. Lew, Joshua B. Tenenbaum, Vikash K. Mansinghka, Armando Solar-Lezama | 17m |
A new variant of knowledge compilation for discrete PPLs uses lazy knowledge compilation to delay compiling subexpressions until forced, and it compiles only… | ||
| nothing matches | ||