Filters








5,417 Hits in 6.5 sec

Improvements to Inference Compilation for Probabilistic Programming in Large-Scale Scientific Simulators [article]

Mario Lezcano Casado, Atilim Gunes Baydin, David Martinez Rubio, Tuan Anh Le, Frank Wood, Lukas Heinrich, Gilles Louppe, Kyle Cranmer, Karen Ng, Wahid Bhimji, Prabhat
2017 arXiv   pre-print
We extend previous work in "inference compilation", which combines universal probabilistic programming and deep learning methods, to large-scale scientific simulators, and introduce a C++ based probabilistic  ...  We successfully use CPProb to interface with SHERPA, a large code-base used in particle physics. Here we describe the technical innovations realized and planned for this library.  ...  CPProb In order to equip large-scale simulation code bases with inference compilation, we developed CPPROB [3] , 1 a probabilistic programming library written in C++14 that allows inference in probabilistic  ... 
arXiv:1712.07901v1 fatcat:h4tef5xshfdsdaslyc2f346jlm

Etalumis: Bringing Probabilistic Programming to Scientific Simulators at Scale [article]

Atılım Güneş Baydin, Lei Shao, Wahid Bhimji, Lukas Heinrich, Lawrence Meadows, Jialin Liu, Andreas Munk, Saeid Naderiparizi, Bradley Gram-Hansen, Gilles Louppe, Mingfei Ma, Xiaohui Zhao, Philip Torr, Victor Lee (+3 others)
2019 arXiv   pre-print
Probabilistic programming languages (PPLs) are receiving widespread attention for performing Bayesian inference in complex generative models.  ...  We demonstrate a Large Hadron Collider (LHC) use-case with the C++ Sherpa simulator and achieve the largest-scale posterior inference in a Turing-complete PPL.  ...  ACKNOWLEDGMENTS The authors would like to acknowledge valuable discussions with Thorsten  ... 
arXiv:1907.03382v2 fatcat:v4yy3ywsqrcr3fblgbkohvn264

Efficient Probabilistic Inference in the Quest for Physics Beyond the Standard Model [article]

Atılım Güneş Baydin, Lukas Heinrich, Wahid Bhimji, Lei Shao, Saeid Naderiparizi, Andreas Munk, Jialin Liu, Bradley Gram-Hansen, Gilles Louppe, Lawrence Meadows, Philip Torr, Victor Lee, Prabhat (+2 others)
2020 arXiv   pre-print
We present a novel probabilistic programming framework that couples directly to existing large-scale simulators through a cross-platform probabilistic execution protocol, which allows general-purpose inference  ...  The execution of existing simulators as probabilistic programs enables highly interpretable posterior inference in the structured model defined by the simulator code base.  ...  Acknowledgments We thank the anonymous reviewers for their constructive comments that helped us improve this paper significantly. This research used resources of the National Energy  ... 
arXiv:1807.07706v5 fatcat:di4pcnop25ekflrxz3oo73gxhq

The frontier of simulation-based inference

Kyle Cranmer, Johann Brehmer, Gilles Louppe
2020 Proceedings of the National Academy of Sciences of the United States of America  
While these simulations provide high-fidelity models, they are poorly suited for inference and lead to challenging inverse problems.  ...  We review the rapidly developing field of simulation-based inference and identify the forces giving additional momentum to the field.  ...  Inference compilation (49) is a preprocessing step for probabilistic programming algorithms, shown in Fig. 1D .  ... 
doi:10.1073/pnas.1912789117 pmid:32471948 fatcat:2dabtkqwtzf6ngy62naz3tvlpy

The frontier of simulation-based inference [article]

Kyle Cranmer, Johann Brehmer, Gilles Louppe
2020 arXiv   pre-print
While these simulations provide high-fidelity models, they are poorly suited for inference and lead to challenging inverse problems.  ...  We review the rapidly developing field of simulation-based inference and identify the forces giving new momentum to the field.  ...  GL is recipient of the ULiège-NRB Chair on Big Data and is thankful for the support of NRB.  ... 
arXiv:1911.01429v3 fatcat:kv32pqap5ne2hkvnekcck4hxkq

Gen: a general-purpose probabilistic programming system with programmable inference

Marco F. Cusumano-Towner, Feras A. Saad, Alexander K. Lew, Vikash K. Mansinghka
2019 Proceedings of the 40th ACM SIGPLAN Conference on Programming Language Design and Implementation - PLDI 2019  
Probabilistic modeling and inference are central to many fields. A key challenge for wider adoption of probabilistic programming languages is designing systems that are both flexible and performant.  ...  This paper introduces Gen, a new probabilistic programming system with novel language constructs for modeling and for end-user customization and optimization of inference.  ...  Compilers have been developed for probabilistic programming languages to improve the performance of inference [6, 21, 50, 51] .  ... 
doi:10.1145/3314221.3314642 dblp:conf/pldi/Cusumano-Towner19 fatcat:bmjwmh7jhjf33gg6dywrdjop7y

Approximate and Probabilistic Computing: Design, Coding, Verification (Dagstuhl Seminar 15491)

Antonio Filieri, Marta Kwiatkowska, Sasa Misailovic, Todd Mytkowicz, Marc Herbstritt
2016 Dagstuhl Reports  
The aim of this seminar was to bring together academic and industrial researchers from the areas of probabilistic model checking, quantitative software analysis, probabilistic programming, and approximate  ...  These complex systems require sophisticated algorithms to deliver accurate answers quickly, at scale, and with energy efficiency, and approximation is often the only way to meet these competing goals.  ...  algorithms, and (2) improve testing of probabilistic inference engines and provide alternative strategies for computing results for some classes of probabilistic inference problems.  ... 
doi:10.4230/dagrep.5.11.151 dblp:journals/dagstuhl-reports/FilieriKMM15 fatcat:dao63covdjhflma6fflt3meus4

Learning Units-of-Measure from Scientific Code

Matthew Danish, Miltiadis Allamanis, Marc Brockschmidt, Andrew Rice, Dominic Orchard
2019 2019 IEEE/ACM 14th International Workshop on Software Engineering for Science (SE4Science)  
However, many users find it onerous to provide units-of-measure information for existing code, even in part.  ...  CamFort is our multi-purpose tool for lightweight analysis and verification of scientific Fortran code.  ...  CamFort provides an inference mode which attempts to infer suitable annotations for variables in a program.  ... 
doi:10.1109/se4science.2019.00013 dblp:conf/icse/DanishABRO19 fatcat:46cylbgcdjbxxo5hktpzzrn3fq

Opportunities and Challenges for Next Generation Computing [article]

Gregory D. Hager, Mark D. Hill, Katherine Yelick
2020 arXiv   pre-print
In the remainder of this paper, we articulate some opportunities and challenges for dramatic performance improvements of both personal to national scale computing, and discuss some "out of the box" possibilities  ...  As a nation, we rely on computing in the design of systems for energy, transportation and defense; and computing fuels scientific discoveries that will improve our fundamental understanding of the world  ...  More radically, direct support for probabilistic values and computation may save conventional processing steps on some problems, such as statistical inference or in machine learning.  ... 
arXiv:2008.00023v1 fatcat:s26receqvjdwzmaeyzvholvnyi

Inferring signaling pathways with probabilistic programming

David Merrell, Anthony Gitter
2020 Bioinformatics  
Our results demonstrate the vast potential for probabilistic programming, and Gen specifically, for biological network inference.  ...  The Gen language, in particular, allows us to customize our inference procedure for biological graphs and ensure efficient sampling.  ...  those resources effectively; and the Gen team (Cusumano-Towner et al., 2019) for designing a uniquely powerful probabilistic programming language.  ... 
doi:10.1093/bioinformatics/btaa861 pmid:33381832 fatcat:isrk5dexkrh7xfvh2pne5mdmnm

Functional probabilistic programming for scalable Bayesian modelling [article]

Jonathan Law, Darren Wilkinson
2019 arXiv   pre-print
Probabilistic programming aims to reduce the barrier to performing Bayesian inference by developing a domain specific language (DSL) for model specification which is decoupled from the parameter inference  ...  This paper introduces functional programming principles which can be used to develop an embedded probabilistic programming language.  ...  Acknowledgements JL is supported by the Engineering and Physical Sciences Research Council, Centre for Doctoral Training in Cloud Computing for Big Data (grant number EP/L015358/1) and Digital Catapult  ... 
arXiv:1908.02062v1 fatcat:2kzvdgrferadxaj7bksnjibqua

BRECCIA: A novel multi-source fusion framework for dynamic geospatial data analysis

D. Sacharny, T. C. Henderson, R. Simmons, A. Mitiche, T. Welker, X. Fan
2017 2017 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI)  
backchain(X) Once the list of probabilities have compiled, the nonlinear probabilistic logic algorithm (described in section III) is executed to calculate the probability for the inferred belief.  ...  Each agent in the BRECCIA system is composed of a backward chaining inference module (see [25] for a formal justification of modularity in BDI programming languages) with a probabilistic logic component  ... 
doi:10.1109/mfi.2017.8170352 dblp:conf/mfi/SacharnyHSMWF17 fatcat:rvo6jnpaqnglllanimybn3p75i

From Ontology Selection and Semantic Web to an Integrated Information System for Food-borne Diseases and Food Safety [chapter]

Xianghe Yan, Yun Peng, Jianghong Meng, Juliana Ruzante, Pina M. Fratamico, Lihan Huang, Vijay Juneja, David S. Needleman
2011 Advances in Experimental Medicine and Biology  
scientists in academia, the food industry, and government agencies; and d) development of a computational model in semantic web for greater adaptability and robustness.  ...  This paper discusses technical aspects in the establishment of a comprehensive food safety information system consisting of the following steps: a) computational collection and compiling publicly available  ...  Data sharing is another large challenge facing the development of FSIRS. Data mining and large scale statistical data analysis is a time-consuming process.  ... 
doi:10.1007/978-1-4419-7046-6_76 pmid:21431616 fatcat:4ujwh2c4ljd53gmyhznlwuoeju

Inferring Signaling Pathways with Probabilistic Programming [article]

David Merrell, Anthony Gitter
2020 arXiv   pre-print
The Gen language, in particular, allows us to customize our inference procedure for biological graphs and ensure efficient sampling.  ...  Our results demonstrate the vast potential for probabilistic programming, and Gen specifically, for biological network inference. Find the full codebase at https://github.com/gitter-lab/ssps  ...  those resources effectively; the Gen team (Cusumano-Towner et al., 2019) for designing a uniquely powerful probabilistic programming language; and the HPN-DREAM challenge organizers for providing experimental  ... 
arXiv:2005.14062v2 fatcat:5zrgjjmxbjc27bou6jw4ffzhf4

Probabilistic programming in Python using PyMC3

John Salvatier, Thomas V. Wiecki, Christopher Fonnesbeck
2016 PeerJ Computer Science  
Probabilistic programming allows for automatic Bayesian inference on user-defined probabilistic models.  ...  PyMC3 is a new open source probabilistic programming framework written in Python that uses Theano to compute gradients via automatic differentiation as well as compile probabilistic programs on-the-fly  ...  Accompanying the rise of probabilistic programming has been a burst of innovation in fitting methods for Bayesian models that represent notable improvement over existing MCMC methods.  ... 
doi:10.7717/peerj-cs.55 fatcat:5dnl6podevb4vgaqfmywwlbcdi
« Previous Showing results 1 — 15 out of 5,417 results