Filters








858 Hits in 7.8 sec

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  
This paper introduces Gen, a new probabilistic programming system with novel language constructs for modeling and for end-user customization and optimization of inference.  ...  repeated computation, enabling speedups from caching; and (iv) a standard inference library that supports custom proposal distributions also written as programs in Gen.  ...  Gen also makes it practical to write inference programs that combine built-in operators for Monte Carlo inference and gradient-based optimization with custom algorithmic proposals and deep inference networks  ... 
doi:10.1145/3314221.3314642 dblp:conf/pldi/Cusumano-Towner19 fatcat:bmjwmh7jhjf33gg6dywrdjop7y

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

Alexander K. Lew, Monica Agrawal, David Sontag, Vikash K. Mansinghka
2020 arXiv   pre-print
PClean is powered by three modeling and inference contributions: (1) a non-parametric model of relational database instances, customizable via probabilistic programs, (2) a sequential Monte Carlo inference  ...  Data cleaning can be naturally framed as probabilistic inference in a generative model, combining a prior distribution over ground-truth databases with a likelihood that models the noisy channel by which  ...  Zia Abedjan, Marco Cusumano-Towner, Raul Castro Fernandez, Cameron Freer, Divya Gopinath, Christina Ji, Tim Kraska, George Matheos, Feras Saad, Michael Stonebraker, Josh Tenenbaum, and Veronica Weiner for  ... 
arXiv:2007.11838v4 fatcat:navjwv7vpfbzhaq4mugkbucvve

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  ...  Picture provides a stochastic scene language that can express generative models for arbitrary 2D/3D scenes, as well as a hierarchy of representation layers for comparing scene hypotheses with observed  ...  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

Compiling Stan to Generative Probabilistic Languages and Extension to Deep Probabilistic Programming [article]

Guillaume Baudart, Javier Burroni, Martin Hirzel, Louis Mandel, Avraham Shinnar
2021 arXiv   pre-print
Stan is a probabilistic programming language that is popular in the statistics community, with a high-level syntax for expressing probabilistic models.  ...  Building on Pyro we extend Stan with support for explicit variational inference guides and deep probabilistic models.  ...  INTRODUCTION Probabilistic Programming Languages (PPLs) are designed to describe probabilistic models and run inference on these models.  ... 
arXiv:1810.00873v5 fatcat:3lcvh6vr6rbxhaszuuuvjpsjuu

Program Analysis of Probabilistic Programs [article]

Maria I. Gorinova
2022 arXiv   pre-print
No single inference algorithm can be used as a probabilistic programming back-end that is simultaneously reliable, efficient, black-box, and general.  ...  Probabilistic programming is a growing area that strives to make statistical analysis more accessible, by separating probabilistic modelling from probabilistic inference.  ...  The program as a whole can then be seen as inferring some variables (those defined in the parameters block) via HMC, and others (those drawn with pseudo-random number generators) via ancestral sampling  ... 
arXiv:2204.06868v1 fatcat:2dbonwruuzaopil4aijdeuz4mi

An Introduction to Probabilistic Programming [article]

Jan-Willem van de Meent, Brooks Paige, Hongseok Yang, Frank Wood
2021 arXiv   pre-print
It not only provides a thorough background for anyone wishing to use a probabilistic programming system, but also introduces the techniques needed to design and build these systems.  ...  Inference requires methods that generate samples by repeatedly evaluating the program.  ...  Most critically, while all of the authors were at Oxford together, three of them were explicitly supported at various times by the DARPA under its Probabilistic Programming for Advanced Machine Learning  ... 
arXiv:1809.10756v2 fatcat:cdcsscxbu5af7fpm5w6mjjiyra

Compiling Stan to generative probabilistic languages and extension to deep probabilistic programming

Guillaume Baudart, Javier Burroni, Martin Hirzel, Louis Mandel, Avraham Shinnar
2021 Proceedings of the 42nd ACM SIGPLAN International Conference on Programming Language Design and Implementation  
Stan is a probabilistic programming language that is popular in the statistics community, with a high-level syntax for expressing probabilistic models.  ...  Building on Pyro we extend Stan with support for explicit variational inference guides and deep probabilistic models.  ...  Acknowledgements The authors are greatful to the following people for their helpful feedback and encouragements: E. Bingham, K. Kate, Y. Mroueh, F. Obermeyer, A. Pauthier, and D. Phan.  ... 
doi:10.1145/3453483.3454058 fatcat:vhgiqkekjrc2xpponmwkojns4q

Website Morphing

John R. Hauser, Glen L. Urban, Guilherme Liberali, Michael Braun
2009 Marketing science (Providence, R.I.)  
"Morphing" involves automatically matching the basic "look and feel" of a website, not just the content, to cognitive styles. We infer cognitive styles from clickstream data with Bayesian updating.  ...  V irtual advisors often increase sales for those customers who find such online advice to be convenient and helpful.  ...  Recent developments in the bandit literature now make it feasible to include switching costs via fast generalized index heuristics (e.g., Dusonchet and Hongler 2006, Jun 2004) .  ... 
doi:10.1287/mksc.1080.0459 fatcat:yeejiwmwcjb3nk4jc5lydt7j5u

World model learning and inference

Karl Friston, Rosalyn J. Moran, Yukie Nagai, Tadahiro Taniguchi, Hiroaki Gomi, Josh Tenenbaum
2021 Neural Networks  
Lastly, we review recent progress in creating human-like intelligence by using novel probabilistic programming languages.  ...  We then describe symbol emergence in the context of probabilistic modelling, as a topic at the frontiers of cognitive robotics.  ...  The authors express special thanks to Kenji Doya (symposium organiser) and Mitsuo Kawato (designated discussant). H.G thanks to J De Havas for editing.  ... 
doi:10.1016/j.neunet.2021.09.011 pmid:34634605 fatcat:ovruawnvivgybmd6keltwmdisy

Inferring Energy Bounds via Static Program Analysis and Evolutionary Modeling of Basic Blocks [article]

Umer Liqat, Zorana Bankovic, Pedro Lopez-Garcia, Manuel V. Hermenegildo
2017 arXiv   pre-print
Our approach divides a program into basic (branchless) blocks and estimates the maximal and minimal energy consumption for each block using an evolutionary algorithm.  ...  In this work we address this challenge from the software point of view and propose a novel parametric approach to estimating tight bounds on the energy consumed by program executions that are practical  ...  We also thank Henk Muller, Principal Technologist, XMOS, for his help with the measurement boards, evaluation platform, benchmarks, and overall support.  ... 
arXiv:1601.02800v2 fatcat:islryqjzineetk2vdjiq4rms2u

Inferring Energy Bounds via Static Program Analysis and Evolutionary Modeling of Basic Blocks [chapter]

Umer Liqat, Zorana Banković, Pedro Lopez-Garcia, Manuel V. Hermenegildo
2018 Lecture Notes in Computer Science  
In this work we address this challenge from the software point of view and propose a novel approach to estimating accurate parametric bounds on the energy consumed by program executions that are practical  ...  Then it combines the obtained values according to the program control flow, using a safe static analysis, to infer functions that give both upper and lower bounds on the energy consumption of the whole  ...  We also thank Henk Muller, Principal Technologist, XMOS, for his help with the measurement boards, evaluation platform, benchmarks, and overall support.  ... 
doi:10.1007/978-3-319-94460-9_4 fatcat:xk772uh3r5dppfig3lp56f3qim

Taming Reflection

Xiaoyu Sun, Li Li, Tegawendé F. Bissyandé, Jacques Klein, Damien Octeau, John Grundy
2021 ACM Transactions on Software Engineering and Methodology  
We propose a new instrumentation-based approach to address this issue in a non-invasive way.  ...  After that, it automatically instruments the app to replace reflective calls with their corresponding Java calls in a traditional paradigm.  ...  The gen keyword specifies that the method generates a new object of type Method (i.e., it is a factory function).  ... 
doi:10.1145/3440033 fatcat:ijkwhqxanfdt5fwbuux44o4sjy

Network Environment Design for Autonomous Cyberdefense [article]

Andres Molina-Markham, Cory Miniter, Becky Powell, Ahmad Ridley
2021 arXiv   pre-print
This paper introduces a novel approach for network environment design and a software framework to address the fundamental problem that network defense cannot be defined as a single game with a simple set  ...  Demonstrating that a specific RL algorithm can be effective for defending a network under certain conditions may not necessarily give insight about the performance of the algorithm when the threats, network  ...  In addition, CybORG's models use finite state machines as opposed to FARLAND's richer probabilistic models via generative programs [9] .  ... 
arXiv:2103.07583v1 fatcat:entwvffalvgphhpgazjtum4djq

Memory-driven computing accelerates genomic data processing [article]

Matthias Becker, Milind Chabbi, Stefanie Warnat-Herresthal, Kathrin Klee, Jonas Schulte-Schrepping, Pawel Biernat, Patrick Guenther, Kevin Bassler, Rocky Craig, Hartmut Schultze, Sharad Singhal, Thomas Ulas (+1 others)
2019 bioRxiv   pre-print
Together with the MDC-inherent solutions for local data privacy, a new compute model can be projected pushing large scale NGS data processing and primary data analytics closer to the edge by directly combining  ...  Even more impressive, pseudoalignment by near-optimal probabilistic RNA-seq quantification (kallisto) was accelerated by more than two orders of magnitude with identical accuracy and indicated 66% reduced  ...  We thank Bill Hayes, Keith Packard, Patrick Demichel, Binoy Arnold, Robert Peter Haddad, Eric Wu, Chris Kirby for supporting us in running the experiments on the HPE infrastructure.  ... 
doi:10.1101/519579 fatcat:6xpx4xzg7zdcrp4tuodk5njcsq

Julia Language in Machine Learning: Algorithms, Applications, and Open Issues [article]

Kaifeng Gao, Jingzhi Tu, Zenan Huo, Gang Mei, Francesco Piccialli, Salvatore Cuomo
2020 arXiv   pre-print
The Julia language is a fast, easy-to-use, and open-source programming language that was originally designed for high-performance computing, which can well balance the efficiency and simplicity.  ...  A simple and efficient programming language could accelerate applications of machine learning in various fields.  ...  For example, Cusumano and Mansinghka [23] proposed a design for a probabilistic programming language called Gen, which is embedded in Julia, which aims to be sufficiently expressive and performant for  ... 
arXiv:2003.10146v1 fatcat:f2ocidpu4rchnokkc46qzrjgyu
« Previous Showing results 1 — 15 out of 858 results