Filters








4,845 Hits in 4.0 sec

Bayesian synthesis of probabilistic programs for automatic data modeling

Feras A. Saad, Marco F. Cusumano-Towner, Ulrich Schaechtle, Martin C. Rinard, Vikash K. Mansinghka
2019 Proceedings of the ACM on Programming Languages (PACMPL)  
We provide a precise formulation of Bayesian synthesis for automatic data modeling that identifies sufficient conditions for the resulting synthesis procedure to be sound.  ...  We present new techniques for automatically constructing probabilistic programs for data analysis, interpretation, and prediction.  ...  ACKNOWLEDGMENTS This research was supported by the DARPA SD2 program (contract FA8750-17-C-0239); grants from the MIT Media Lab, the Harvard Berkman Klein Center Ethics and Governance of AI Fund, and the  ... 
doi:10.1145/3290350 fatcat:xemazron3rg65nvmab2rdcgyei

Time Series Structure Discovery via Probabilistic Program Synthesis [article]

Ulrich Schaechtle, Feras Saad, Alexey Radul, Vikash Mansinghka
2017 arXiv   pre-print
This paper shows how to extend ABCD by formulating it in terms of probabilistic program synthesis.  ...  The final probabilistic program is written in under 70 lines of probabilistic code in Venture.  ...  learning as probabilistic program synthesis Our objective in probabilistic program synthesis for Bayesian structure learning is to learn a symbolic representation of a probabilistic model program, by  ... 
arXiv:1611.07051v3 fatcat:6pehz3plczhvvo3qadflpsmsxu

Approximate Bayesian Image Interpretation using Generative Probabilistic Graphics Programs [article]

Vikash K. Mansinghka, Tejas D. Kulkarni, Yura N. Perov, Joshua B. Tenenbaum
2013 arXiv   pre-print
Generative probabilistic graphics programs consist of a stochastic scene generator, a renderer based on graphics software, a stochastic likelihood model linking the renderer's output and the data, and  ...  Each of the probabilistic graphics programs we present relies on under 20 lines of probabilistic code, and supports accurate, approximately Bayesian inferences about ambiguous real-world images.  ...  Acknowledgments We are grateful to Keith Bonawitz and Eric Jonas for preliminary work exploring the feasibility of CAPTCHA breaking in Church, and to Seth Teller, Bill Freeman, Ted Adelson, Michael James  ... 
arXiv:1307.0060v1 fatcat:2rdpf4fn4fhs3ccwuzzkbv2wqi

Automatic Sampler Discovery via Probabilistic Programming and Approximate Bayesian Computation [chapter]

Yura Perov, Frank Wood
2016 Lecture Notes in Computer Science  
We describe an approach to automatic discovery of samplers in the form of human interpretable probabilistic programs.  ...  We formulate a Bayesian approach to this problem by specifying an adaptor grammar prior over probabilistic program code, and use approximate Bayesian computation to learn a program whose execution generates  ...  Ultimately, the artificial general intelligence machinery can use such approach to synthesise and update the model of the world in the form of probabilistic programs.  ... 
doi:10.1007/978-3-319-41649-6_27 fatcat:3jbkfa2t7bhubph2h6t6wl4z6e

Evidence Synthesis for Decision Making 6

Sofia Dias, Alex J. Sutton, Nicky J. Welton, A. E. Ades
2013 Medical decision making  
Software suitable for transferring data between different packages, and software that provides a userfriendly interface for integrated software platforms, offering investigators a flexible way of examining  ...  The first is evidence synthesis by Bayesian posterior estimation and posterior sampling where other parameters of the costeffectiveness model can be incorporated into the same software platform.  ...  versions of this paper.  ... 
doi:10.1177/0272989x13487257 pmid:23804510 pmcid:PMC3704202 fatcat:ltgu32bpjbau7au6k2azjilk4e

Picture: A probabilistic programming language for scene perception

Tejas D Kulkarni, Pushmeet Kohli, Joshua B Tenenbaum, Vikash Mansinghka
2015 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
Here we present Picture, a probabilistic programming language for scene understanding that allows researchers to express complex generative vision models, while automatically solving them using fast general-purpose  ...  Recent progress on probabilistic modeling and statistical learning, coupled with the availability of large training datasets, has led to remarkable progress in computer vision.  ...  Acknowledgements We thank Thomas Vetter for giving us access to the Basel face model. T. Kulkarni was graciously supported by the Leventhal Fellowship.  ... 
doi:10.1109/cvpr.2015.7299068 dblp:conf/cvpr/KulkarniKTM15 fatcat:kwyidcyug5esxiu5jykt6knim4

Applications of Probabilistic Programming (Master's thesis, 2015) [article]

Yura N Perov
2020 arXiv   pre-print
We implement this approach in the probabilistic programming system Anglican, and show that for that model data-driven proposals provide significant performance improvements.  ...  In Chapter 1 we provide a brief introduction to probabilistic programming. In Chapter 2 we present an approach to automatic discovery of samplers in the form of probabilistic programs.  ...  The ultimate goal is to find a way to generate such data-driven proposals automatically, given a generative model in the form of a probabilistic program.  ... 
arXiv:1606.00075v2 fatcat:bwargqfv25fjplh53spm7aitbe

Efficient synthesis of probabilistic programs

Aditya V. Nori, Sherjil Ozair, Sriram K. Rajamani, Deepak Vijaykeerthy
2015 Proceedings of the 36th ACM SIGPLAN Conference on Programming Language Design and Implementation - PLDI 2015  
We show how to automatically synthesize probabilistic programs from real-world datasets.  ...  A core difficulty in synthesizing probabilistic programs is computing the likelihood L(P | D) of a candidate program P generating data D.  ...  Acknowledgments We thank Timon Gehr and Martin Vechev for their help with fixing errors and improving the presentation of this paper.  ... 
doi:10.1145/2737924.2737982 dblp:conf/pldi/NoriORV15 fatcat:6k3nkkbzfra5vplekzxot3rsoi

Efficient synthesis of probabilistic programs

Aditya V. Nori, Sherjil Ozair, Sriram K. Rajamani, Deepak Vijaykeerthy
2015 SIGPLAN notices  
We show how to automatically synthesize probabilistic programs from real-world datasets.  ...  A core difficulty in synthesizing probabilistic programs is computing the likelihood L(P | D) of a candidate program P generating data D.  ...  Acknowledgments We thank Timon Gehr and Martin Vechev for their help with fixing errors and improving the presentation of this paper.  ... 
doi:10.1145/2813885.2737982 fatcat:4zrdv2tnhvb67dowwxnpflj6ii

AutoBayes: a system for generating data analysis programs from statistical models

BERND FISCHER, JOHANN SCHUMANN
2003 Journal of functional programming  
In this paper, we describe AutoBayes, a program synthesis system for the generation of data analysis programs from statistical models.  ...  It is well-suited for tasks like estimating best-fitting model parameters for the given data. Here, we describe AutoBayes's system architecture, in particular the schema-guided synthesis kernel.  ...  Acknowledgements: Wray Buntine and Tom Pressburger contributed much to the initial development of AutoBayes. Grigore Rosu implemented the test data generator and the graph visualization.  ... 
doi:10.1017/s0956796802004562 fatcat:tosv54uvizdqlbpom7y6gtwuxu

Learning Probabilistic Programs [article]

Yura N. Perov, Frank D. Wood
2014 arXiv   pre-print
We develop a technique for generalising from data in which models are samplers represented as program text.  ...  We also introduce a new notion of probabilistic program compilation and show how the same machinery might be used in the future to compile probabilistic programs for efficient reusable predictive inference  ...  Government is authorized to reproduce and distribute reprints for Governmental purposes notwithstanding any copyright notation heron.  ... 
arXiv:1407.2646v1 fatcat:gzapxdw7xvbk7m2eodqjjjpgwm

Machine Learning and Model Checking Join Forces (Dagstuhl Seminar 18121)

Nils Jansen, Joost-Pieter Katoen, Pusmeet Kohli, Jan Kretinsky, Michael Wagner
2018 Dagstuhl Reports  
This report documents the program and the outcomes of Dagstuhl Seminar 18121 "Machine Learning and Model Checking Join Forces".  ...  This Dagstuhl Seminar brought together researchers working in the fields of machine learning and model checking.  ...  Among other applications, NWR-based MDP reductions can be seen as a pre-processing of MDPs before model checking or as a way to reduce the number of experiments required to obtain a good approximation  ... 
doi:10.4230/dagrep.8.3.74 dblp:journals/dagstuhl-reports/JansenKKK18 fatcat:225qaztsujhgxpclahyf4wm7qe

Development of an internet based system for modeling biotin metabolism using Bayesian networks

Jinglei Zhou, Dong Wang, Vicki Schlegel, Janos Zempleni
2011 Computer Methods and Programs in Biomedicine  
In this paper, we report the development of BiotinNet, an internet based program that uses Bayesian networks to integrate published data on various aspects of biotin metabolism.  ...  The importance of biotin for human health has been under-appreciated but there is plenty of opportunity for future research with great importance for human health.  ...  The Bayesian networks are particular attractive for this task because of the flexibility of using Bayesian graphical models for knowledge synthesis.  ... 
doi:10.1016/j.cmpb.2011.02.004 pmid:21356565 pmcid:PMC3132571 fatcat:qefjojajl5crlklhrtpxfa6mv4

Will domain-specific code synthesis become a silver bullet?

W. Buntine, P. Norvig, J. Van Baalen, D. Spiegelhalter, A. Thomas
1998 IEEE Intelligent Systems and their Applications  
A Doodle for a model of growth patterns of rats.  ...  Software is increasingly difficult to develop; to be cost-effective, most future software devel-We plan to demonstrate this data-understanding program-synthesis fool in fhe synthesis of software for onboard  ...  It is not suitable for people who want to fit standard regression models to data and can get by with a standard $500 statistics package.  ... 
doi:10.1109/5254.671084 fatcat:5tncsmbhirebrarcd6s7vh6d4q

Data-Driven Synthesis of Full Probabilistic Programs [chapter]

Sarah Chasins, Phitchaya Mangpo Phothilimthana
2017 Lecture Notes in Computer Science  
Unfortunately, writing a PPL program by hand can be difficult for non-experts, requiring extensive knowledge of statistics and deep insights into the data.  ...  We introduce a data-guided approach to the program mutation stage of simulated annealing; this innovation allows our tool to scale to synthesizing complete probabilistic programs from scratch.  ...  Department of Energy, Office of Science, Office of Basic Energy Sciences Energy Frontier Research Centers program under Award Number FOA-0000619, and grants from DARPA FA8750-14-C-0011 and DARPA FA8750  ... 
doi:10.1007/978-3-319-63387-9_14 fatcat:idnaf7svwrc2thcmndstn6gjxm
« Previous Showing results 1 — 15 out of 4,845 results