PClean: Bayesian Data Cleaning at Scale with Domain-Specific Probabilistic Programming

Alexander Lew, Monica Agrawal, David Sontag, Vikash Mansinghka
Proceedings of The 24th International Conference on Artificial Intelligence and Statistics, PMLR 130:1927-1935, 2021.

Abstract

Data cleaning is naturally framed as probabilistic inference in a generative model of ground-truth data and likely errors, but the diversity of real-world error patterns and the hardness of inference make Bayesian approaches difficult to automate. We present PClean, a probabilistic programming language (PPL) for leveraging dataset-specific knowledge to automate Bayesian cleaning. Compared to general-purpose PPLs, PClean tackles a restricted problem domain, enabling three modeling and inference innovations: (1) a non-parametric model of relational database instances, which users’ programs customize; (2) a novel sequential Monte Carlo inference algorithm that exploits the structure of PClean’s model class; and (3) a compiler that generates near-optimal SMC proposals and blocked-Gibbs rejuvenation kernels based on the user’s model and data. We show empirically that short (< 50-line) PClean programs can: be faster and more accurate than generic PPL inference on data-cleaning benchmarks; match state-of-the-art data-cleaning systems in terms of accuracy and runtime (unlike generic PPL inference in the same runtime); and scale to real-world datasets with millions of records.

Cite this Paper

BibTeX
@InProceedings{pmlr-v130-lew21a, title = { PClean: Bayesian Data Cleaning at Scale with Domain-Specific Probabilistic Programming }, author = {Lew, Alexander and Agrawal, Monica and Sontag, David and Mansinghka, Vikash}, booktitle = {Proceedings of The 24th International Conference on Artificial Intelligence and Statistics}, pages = {1927--1935}, year = {2021}, editor = {Banerjee, Arindam and Fukumizu, Kenji}, volume = {130}, series = {Proceedings of Machine Learning Research}, month = {13--15 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v130/lew21a/lew21a.pdf}, url = {https://proceedings.mlr.press/v130/lew21a.html}, abstract = { Data cleaning is naturally framed as probabilistic inference in a generative model of ground-truth data and likely errors, but the diversity of real-world error patterns and the hardness of inference make Bayesian approaches difficult to automate. We present PClean, a probabilistic programming language (PPL) for leveraging dataset-specific knowledge to automate Bayesian cleaning. Compared to general-purpose PPLs, PClean tackles a restricted problem domain, enabling three modeling and inference innovations: (1) a non-parametric model of relational database instances, which users’ programs customize; (2) a novel sequential Monte Carlo inference algorithm that exploits the structure of PClean’s model class; and (3) a compiler that generates near-optimal SMC proposals and blocked-Gibbs rejuvenation kernels based on the user’s model and data. We show empirically that short (< 50-line) PClean programs can: be faster and more accurate than generic PPL inference on data-cleaning benchmarks; match state-of-the-art data-cleaning systems in terms of accuracy and runtime (unlike generic PPL inference in the same runtime); and scale to real-world datasets with millions of records. } }
Endnote
%0 Conference Paper %T PClean: Bayesian Data Cleaning at Scale with Domain-Specific Probabilistic Programming %A Alexander Lew %A Monica Agrawal %A David Sontag %A Vikash Mansinghka %B Proceedings of The 24th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2021 %E Arindam Banerjee %E Kenji Fukumizu %F pmlr-v130-lew21a %I PMLR %P 1927--1935 %U https://proceedings.mlr.press/v130/lew21a.html %V 130 %X Data cleaning is naturally framed as probabilistic inference in a generative model of ground-truth data and likely errors, but the diversity of real-world error patterns and the hardness of inference make Bayesian approaches difficult to automate. We present PClean, a probabilistic programming language (PPL) for leveraging dataset-specific knowledge to automate Bayesian cleaning. Compared to general-purpose PPLs, PClean tackles a restricted problem domain, enabling three modeling and inference innovations: (1) a non-parametric model of relational database instances, which users’ programs customize; (2) a novel sequential Monte Carlo inference algorithm that exploits the structure of PClean’s model class; and (3) a compiler that generates near-optimal SMC proposals and blocked-Gibbs rejuvenation kernels based on the user’s model and data. We show empirically that short (< 50-line) PClean programs can: be faster and more accurate than generic PPL inference on data-cleaning benchmarks; match state-of-the-art data-cleaning systems in terms of accuracy and runtime (unlike generic PPL inference in the same runtime); and scale to real-world datasets with millions of records.
APA
Lew, A., Agrawal, M., Sontag, D. & Mansinghka, V.. (2021). PClean: Bayesian Data Cleaning at Scale with Domain-Specific Probabilistic Programming . Proceedings of The 24th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 130:1927-1935 Available from https://proceedings.mlr.press/v130/lew21a.html.

Related Material