检索结果-内蒙古大学图书馆

41st ACM SIGPLAN Conference on programming Language Design and Implementation (PLDI)

作者： Gehr, Timon Steffen, Samuel Vechev, Martin Swiss Fed Inst Technol Zurich Switzerland

ISBN: (纸本)9781450376136

We present lambda PSI, the first probabilistic programming language and system that supports higher-order exact inference for probabilistic programs with first-class functions, nested inference and discrete, continuous and mixed random variables. lambda PSI's solver is based on symbolic reasoning and computes the exact distribution represented by a program. We show that lambda PSI is practically effective-it automatically computes exact distributions for a number of interesting applications, from rational agents to information theory, many of which could so far only be handled approximately.

关键词： probabilistic programming Exact Higher-order

来源：评论

学校读者我要写书评

暂无评论

Slicing probabilistic Programs 14

Slicing Probabilistic Programs

引用

35th ACM SIGPLAN Conference on programming Language Design and Implementation (PLDI)

作者： Hur, Chung-Kil Nori, Aditya V. Rajamani, Sriram K. Samuel, Selva Seoul Natl Univ Seoul 151 South Korea

ISBN: (纸本)9781450327848

probabilistic programs use familiar notation of programming languages to specify probabilistic models. Suppose we are interested in estimating the distribution of the return expression r of a probabilistic program P. We are interested in slicing the probabilistic program P and obtaining a simpler program SLI (P) which retains only those parts of P that are relevant to estimating r, and elides those parts of P that are not relevant to estimating r. We desire that the SLI transformation be both correct and efficient. By correct, we mean that P and SLI (P) have identical estimates on r. By efficient, we mean that estimation over SLI (P) be as fast as possible. We show that the usual notion of program slicing, which traverses control and data dependencies backward from the return expression r, is unsatisfactory for probabilistic programs, since it produces incorrect slices on some programs and sub-optimal ones on others. Our key insight is that in addition to the usual notions of control dependence and data dependence that are used to slice non-probabilistic programs, a new kind of dependence called observe dependence arises naturally due to observe statements in probabilistic programs. We propose a new definition of SLI (P) which is both correct and efficient for probabilistic programs, by including observe dependence in addition to control and data dependences for computing slices. We prove correctness mathematically, and we demonstrate efficiency empirically. We show that by applying the SLI transformation as a pre-pass, we can improve the efficiency of probabilistic inference, not only in our own inference tool R2, but also in other systems for performing inference such as Church and Infer. NET.

关键词： probabilistic programming Program Slicing Bayesian Reasoning

来源：评论

学校读者我要写书评

暂无评论

Debugging probabilistic Programs 1

Debugging Probabilistic Programs

引用

4th ACM SIGPLAN International Workshop on Libraries, Languages, and Compilers for Array programming (ARRAY) / 1st ACM SIGPLAN International Workshop on Machine Learning and programming Languages (MAPL)

作者： Nandi, Chandrakana Sampson, Adrian Mytkowicz, Todd McKinley, Kathryn S. Univ Washington Seattle WA 98195 USA Cornell Univ Ithaca NY USA Microsoft Res Redmond WA USA Google Mountain View CA USA

ISBN: (纸本)9781450350716

Many applications compute with estimated and uncertain data. While advances in probabilistic programming help developers build such applications, debugging them remains extremely challenging. New types of errors in probabilistic programs include 1) ignoring dependencies and correlation between random variables and in training data, 2) poorly chosen inference hyper-parameters, and 3) incorrect statistical models. A partial solution to prevent these errors in some languages forbids developers from explicitly invoking inference. While this prevents some dependence errors, it limits composition and control over inference, and does not guarantee absence of other types of errors. This paper presents the FLEXI programming model which supports constructs for invoking inference in the language and reusing the results in other statistical computations. We define a novel formalism for inference with a Decorated Bayesian Network and present a tool, DePP, that analyzes this representation to identify the above errors. We evaluate DePP on a range of prototypical examples to show how it helps developers to detect errors.

关键词： probabilistic programming debugging program analysis statistical inference

来源：评论

学校读者我要写书评

暂无评论

Sustainability at Scale: Towards Bridging the Intention-Behavior Gap with Sustainable Recommendations 18

Sustainability at Scale: Towards Bridging the Intention-Beha...

引用

12th ACM Conference on Recommender Systems (RecSys)

作者： Tomkins, Sabina Isley, Steven London, Ben Getoor, Lise UC Santa Cruz Santa Cruz CA 95064 USA Amazon Seattle WA USA

ISBN: (纸本)9781450359016

Finding sustainable products and evaluating their claims is a significant barrier facing sustainability-minded customers. Tools that reduce both these burdens are likely to boost the sale of sustainable products. However, it is difficult to determine the sustainability characteristics of these products - there are a variety of certifications and definitions of sustainability, and quality labeling requires input from domain experts. In this paper, we propose a flexible probabilistic framework that uses domain knowledge to identify sustainable products and customers, and uses these labels to predict customer purchases. We evaluate our approach on grocery items from the Amazon catalog. Our proposed approach outperforms established recommender system models in predicting future purchases while jointly inferring sustainability scores for customers and products.

关键词： Sustainability probabilistic programming

来源：评论

学校读者我要写书评

暂无评论

Operational Planning using Climatological Observations for Maritime Prediction and Analysis Support Service (COMPASS) 11

Operational Planning using Climatological Observations for M...

引用

Conference on Modeling and Simulation for Defense Systems and Applications XI

作者： O'Connor, Alison Kirtman, Benjamin Harrison, Scott Gorman, Joe Charles River Analyt 625 Mt Auburn St Cambridge MA 02318 USA Univ Miami Rosential Sch Marine & Atmospher Sci 4600 Rickenbacker Causeway Miami FL 33149 USA

ISBN: (纸本)9781510600898

The US Navy faces several limitations when planning operations in regard to forecasting environmental conditions. Currently, mission analysis and planning tools rely heavily on short-term (less than a week) forecasts or long-term statistical climate products. However, newly available data in the form of weather forecast ensembles provides dynamical and statistical extended-range predictions that can produce more accurate predictions if ensemble members can be combined correctly. Charles River Analytics is designing the Climatological Observations for Maritime Prediction and Analysis Support Service (COMPASS), which performs data fusion over extended-range multi-model ensembles, such as the North American Multi-Model Ensemble (NMME), to produce a unified forecast for several weeks to several seasons in the future. We evaluated thirty years of forecasts using machine learning to select predictions for an all-encompassing and superior forecast that can be used to inform the Navy's decision planning process.

关键词： Extended-range forecasting probabilistic forecasting multi-model ensembles climate predictability machine learning probabilistic programming Figaro programming language

来源：评论

学校读者我要写书评

暂无评论

Relaxations for probabilistically constrained stochastic programming problems: review and extensions

引用

Annals of Operations Research 2018年 1-22页

作者： Lejeune, Miguel A. Prékopa, A. Department of Decision Sciences George Washington University Washington DC United States Rutgers University Piscataway NJ United States

We consider probabilistically constrained stochastic programming problems, in which the random variables are in the right-hand sides of the stochastic inequalities defining the joint chance constraints. Problems of that kind arise in a variety of contexts, and are particularly difficult to solve for random variables with continuous joint distributions, because the calculation of the cumulative distribution function and its gradient values involves numerical integration and/or simulation in large dimensional spaces. We revisit known and provide new relaxations extensions to various probability bounding schemes that permit to approximate the feasible set of joint probabilistic constraints. The derived mathematical formulations relax the requirement to handle large multivariate cumulative distribution functions and involve instead the computation of marginal and bivariate cumulative distribution functions. We analyze the convexity of and computational challenges posed by the inferred relaxations © 2018 Springer Science+Business Media, LLC, part of Springer Nature

关键词： Chance constraints probabilistic programming Relaxation and approximation problems Reliability Stochastic programming

来源：评论

学校读者我要写书评

暂无评论

Demonstrating Principled Uncertainty Modeling for Recommender Ecosystems with RecSim NG 20

Demonstrating Principled Uncertainty Modeling for Recommende...

引用

14th ACM Conference on Recommender Systems (RECSYS)

作者： Mladenov, Martin Hsu, Chih-wei Jain, Vihan Ie, Eugene Colby, Christopher Mayoraz, Nicolas Pham, Hubert Tran, Dustin Vendrov, Ivan Boutilier, Craig Google Res Mountain View CA 94043 USA

We develop RECSIM NG, a probabilistic platform that supports natural, concise specification and learning of models for multi-agent recommender systems simulation. RECSIM NG is a scalable, modular, differentiable simul... 详细信息

ISBN: (纸本)9781450375832

关键词： probabilistic programming Latent Variable Models Reinforcement Learning

来源：评论

学校读者我要写书评

暂无评论

Unifying Logic and Probability: A New Dawn for AI?

Unifying Logic and Probability: A New Dawn for AI?

引用

15th International Conference on Information Processing and Management of Uncertainty in Knowledge-based Systems (IPMU)

作者： Russell, Stuart Univ Calif Berkeley Berkeley CA 94720 USA

ISBN: (纸本)9783319087955;9783319087948

Logic and probability theory are two of the most important branches of mathematics and each has played a significant role in artificial intelligence (AI) research. Beginning with Leibniz, scholars have attempted to unify logic and probability. For "classical" AI, based largely on first-order logic, the purpose of such a unification is to handle uncertainty and facilitate learning from real data;for "modern" AI, based largely on probability theory, the purpose is to acquire formal languages with sufficient expressive power to handle complex domains and incorporate prior knowledge. This paper provides a brief summary of an invited talk describing efforts in these directions, focusing in particular on open-universe probability models that allow for uncertainty about the existence and identity of objects.

关键词： first-order logic probability probabilistic programming Bayesian logic machine learning

来源：评论

学校读者我要写书评

暂无评论

Generative Datalog with Continuous Distributions 20

Generative Datalog with Continuous Distributions

引用

39th ACM SIGMOD-SIGACT-SIGAI Symposium on Principles of Database Systems (PODS)

作者： Grohe, Martin Kaminski, Benjamin Lucien Katoen, Joost-Pieter Lindner, Peter Rhein Westfal TH Aachen Aachen Germany UCL London England

ISBN: (纸本)9781450371087

Arguing for the need to combine declarative and probabilistic programming, Barany et al. (TODS 2017) recently introduced a probabilistic extension of Datalog as a "purely declarative probabilistic programming language." We revisit this language and propose a more foundational approach towards defining its semantics. It is based on standard notions from probability theory known as stochastic kernels and Markov processes. This allows us to extend the semantics to continuous probability distributions, thereby settling an open problem posed by Barany et al. We show that our semantics is fairly robust, allowing both parallel execution and arbitrary chase orders when evaluating a program. We cast our semantics in the framework of infinite probabilistic databases (Grohe and Lindner, ICDT 2020), and we show that the semantics remains meaningful even when the input of a probabilistic Datalog program is an arbitrary probabilistic database.

关键词： Datalog probabilistic Databases Generative Datalog Measure Theory Stochastic Kernels probabilistic programming

来源：评论

学校读者我要写书评

暂无评论

Hybrid-AI Approach to Health Monitoring of Vehicle Control System 70

Hybrid-AI Approach to Health Monitoring of Vehicle Control S...

引用

70th Annual Reliability and Maintainability Symposium (RAMS)

作者： Lu, Kenneth Hiett, Margarita Cross, Ernest Vincent Reposa, Michael Kain, Aron Davis, Erik Charles River Analyt 625 Mt Auburn St Cambridge MA 02138 USA BH Technol LLC 26 Firemens Mem Dr Pomona NY 10970 USA

ISBN: (纸本)9798350307696

Advances in Artificial Intelligence and Machine Learning AI/ML have demonstrated enormous potential in improving and optimizing condition-based maintenance processes. In this paper, we present novel research that leverages the power of probabilistic programming and hybrid-AI that combines domain knowledge with data to create an effective analytic capability that monitors (in real-time) the health and status of a Robotic Combat Vehicle, next generation prototype ground vehicle for the US Army.

关键词： probabilistic programming Diagnostics Robotic Combat Vehicle Sensor Fusion

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：