PLSE Seminar: Feras Saad: SPPL: Probabilistic Programming with Fast Exact Symbolic Inference
, MIT CSAIL
Date: Thursday, August 05, 2021
Time: 11:00 AM to 12:00 PM Note: all times are in the Eastern Time Zone
Contact: Alexander D Renda, firstname.lastname@example.org
Relevant URL: http://projects.csail.mit.edu/pl/seminars.html
Speaker URL: http://fsaad.mit.edu
TALK: PLSE Seminar: Feras Saad: SPPL: Probabilistic Programming with Fast Exact Symbolic Inference
Abstract: We present the Sum-Product Probabilistic Language (SPPL), a new probabilistic programming language that automatically delivers exact solutions to a broad range of probabilistic inference queries. SPPL translates probabilistic programs into sum-product expressions, a new symbolic representation and associated semantic domain that supports mixed-type distributions, numeric transformations, logical formulas, and pointwise and set-valued constraints. We formalize SPPL via a novel translation strategy from probabilistic programs to sum-product expressions and give sound exact algorithms for conditioning on and computing probabilities of events. SPPL imposes a collection of restrictions on probabilistic programs to ensure they can be translated into sum-product expressions, which allow the system to leverage new techniques for improving the scalability of translation and inference by automatically exploiting probabilistic structure.
We implement a prototype of SPPL with a modular architecture and evaluate it on benchmarks the system targets, showing that it obtains up to 3500x speedups over state-of-the-art symbolic systems on tasks such as verifying the fairness of decision tree classifiers, smoothing hidden Markov models, conditioning transformed random variables, and computing rare event probabilities.
Speaker bio: Feras is a PhD student at the MIT Probabilistic Computing Project, advised by Dr. Vikash Mansinghka. His research interests include developing probabilistic languages for Bayesian model discovery and data analysis, assessing the accuracy of high-dimensional sampling algorithms, well as the foundations of random variate generation.
Paper URL: https://dl.acm.org/doi/10.1145/3453483.3454078
Code URL: https://github.com/probcomp/sppl
Programming Languages & Software
Created by Alexander D Renda at Friday, July 30, 2021 at 3:42 PM.