检索结果-内蒙古大学图书馆

On the Applicability of probabilistic programming languages for Causal Activity Recognition

KUNSTLICHE INTELLIGENZ 2019年第4期33卷 389-399页

作者： Luedtke, Stefan Popko, Maximilian Kirste, Thomas Univ Rostock Inst Comp Sci Rostock Germany

Recognizing causal activities of human protagonists, and jointly inferring context information like location of objects and agents from noisy sensor data is a challenging task. Causal models can be used, which describe the activity structure symbolically, e.g. by precondition-effect actions. Recently, probabilistic programming languages (PPLs) arose as an abstraction mechanism that allow to concisely define probabilistic models by a general-purpose programming language, and provide off-the-shelf, general-purpose inference algorithms. In this paper, we empirically investigate whether PPLs provide a feasible alternative for implementing causal models for human activity recognition, by comparing the performance of three different PPLs (Anglican, WebPPL and Figaro) on a multi-agent scenario. We find that PPLs allow to concisely express causal models, but general-purpose inference algorithms that are typically implemented in PPLs are outperformed by an application-specific inference algorithm by orders of magnitude. Still, PPLs can be a valuable tool for developing probabilistic models, due to their expressiveness and simple applicability.

关键词： Bayesian filtering Causal model probabilistic programming language Anglican WebPPL Figaro Particle filter

来源：评论

学校读者我要写书评

暂无评论

probabilistic models for the guarded command language

引用

SCIENCE OF COMPUTER programming 1997年第2-3期28卷 171-192页

作者： He, JF Seidel, K McIver, A Oxford University Computing Laboratory Programming Research Group 11 Keble. Road Oxford OX1 3QD UK

The two models presented in this paper provide two different semantics for an extension of Dijkstra's language of guarded commands. The extended language has an additional operator, namely probabilistic choice, which makes it possible to express randomized algorithms. An earlier model by Claire Jones included probabilistic choice but not non-determinism, which meant that it could not be used for the development of algorithms from specifications. Our second model is built on top of Claire Jones' model, using a general method of extending a probabilistic cpo to one which also contains non-determinism. The first model was constructed from scratch, as it were, guided only by the desire for certain algebraic properties of the language constructs, which we found lacking in the second model. We compare and contrast the properties of the two models both by giving examples and by constructing mappings between them and the non-probabilistic model. On the basis of this comparison we argue that, in general, the first model is preferable to the second. (C) 1997 Elsevier Science B.V.

关键词： probabilistic programming language semantics algebraic laws

来源：评论

学校读者我要写书评

暂无评论

Operator Algebras and the Operational Semantics of probabilistic languages

引用

ELECTRONIC NOTES IN THEORETICAL COMPUTER SCIENCE 2006年第1期161卷 131-150页

作者： Di Pierro, Alessandra Wiklicky, Herbert Univ Pisa Dipartimento Informat Pisa Italy Imperial Coll London Dept Comp London England

We investigate the construction of linear operators representing the semantics of probabilistic programming languages expressed via probabilistic transition systems. Finite transition relations, corresponding to finite automata, can easily be represented by finite dimensional matrices;for the infinite case we need to consider an appropriate generalisation of matrix algebras. We argue that C*-algebras, or more precisely Approximately Finite (or AF) algebras, provide a sufficiently rich mathematical structure for modelling probabilistic processes. We show how to construct for a given probabilistic language a unique AF algebra A and how to represent the operational semantics of processes within this framework: finite computations correspond directly to operators in A, while infinite processes are represented by elements in the so-called strong closure of this algebra.

关键词： Linear operator C*-algebra AF-algebra operational semantics probabilistic programming language

来源：评论

学校读者我要写书评

暂无评论

Bayesian Statistical Model of Item Response Theory in Observer Studies of Radiologists

引用

ACADEMIC RADIOLOGY 2020年第3期27卷 E45-E54页

作者： Nishio, Mizuho Akasaka, Thai Sakamoto, Ryo Togashi, Kaori Kyoto Univ Dept Diagnost Imaging & Nucl Med Grad Sch Med Sakyo Ku 54 Kawahara Cho Kyoto 6068507 Japan Kyoto Univ Hosp Preempt Med & Lifestyle Related Dis Res Ctr Kyoto Japan Osaka Red Cross Hosp Dept Radiol Osaka Japan

Rationale and Objectives: The purpose of this study was to validate a Bayesian statistical model of item response theory (IRT). IRT was used to evaluate a new modality (temporal subtraction, TS) in observer studies of radiologists, compared with a conventional modality (computed tomography). Materials and Methods: From previously published papers, we obtained two datasets of clinical observer studies of radiologists. Those studies used a multi-reader and multi-case paradigm to evaluate radiologists' detection abilities, primarily to determine if TS could enhance the detectability of bone metastasis or brain infarctions. We applied IRT to these studies' datasets using Stan software. Before applying IRT, the radiologists' responses were recorded as binaries for each case (1 = correct, 0 = incorrect). Effect of TS on detectability was evaluated by using our IRT model and calculating the 95% credible interval of the effect. Results: The mean, median, and 95% credible interval of the effect of TS were 0.913, 0.885, and 0.243-1.745 for the bone metastasis detection, and 2.524, 2.50, and 1.827-3.310, for the brain infarction detection. For both detection studies, the 95% credible intervals of the effect of TS did not include zero, indicating that TS significantly improved diagnostic ability. Conclusion: Judgments based on the present study results were compatible with the two previous studies. Our study results demonstrated that the Bayesian statistical model of IRT could judge a new modality's usefulness.

关键词： Bayesian statistical model item response theory observer study probabilistic programming language p-value

来源：评论

学校读者我要写书评

暂无评论

Bayesian multidimensional nominal response model for observer study of radiologists

引用

JAPANESE JOURNAL OF RADIOLOGY 2023年第4期41卷 449-455页

作者： Nishio, Mizuho Kobayashi, Daigo Matsuo, Hidetoshi Urase, Yasuyo Nishioka, Eiko Murakami, Takamichi Kobe Univ Grad Sch Med Dept Radiol 7-5-2 Kusunoki ChoChuo Ku Kobe 6500017 Japan

Purpose This study proposes a Bayesian multidimensional nominal response model (MD-NRM) to statistically analyze the nominal response of multiclass classifications. Materials and methods First, for MD-NRM, we extended the conventional nominal response model to achieve stable convergence of the Bayesian nominal response model and utilized multidimensional ability parameters. We then applied MD-NRM to a 3-class classification problem, where radiologists visually evaluated chest X-ray images and selected their diagnosis from one of the three classes. The classification problem consisted of 150 cases, and each of the six radiologists selected their diagnosis based on a visual evaluation of the images. Consequently, 900 (= 150 x 6) nominal responses were obtained. In MD-NRM, we assumed that the responses were determined by the softmax function, the ability of radiologists, and the difficulty of images. In addition, we assumed that the multidimensional ability of one radiologist were represented by a 3 x 3 matrix. The latent parameters of the MD-NRM (ability parameters of radiologists and difficulty parameters of images) were estimated from the 900 responses. To implement Bayesian MD-NRM and estimate the latent parameters, a probabilistic programming language (Stan, version 2.21.0) was used. Results For all parameters, the Rhat values were less than 1.10. This indicates that the latent parameters of the MD-NRM converged successfully. Conclusion The results show that it is possible to estimate the latent parameters (ability and difficulty parameters) of the MD-NRM using Stan. Our code for the implementation of the MD-NRM is available as open source.

关键词： Item response theory Nominal response model probabilistic programming language Chest X-ray

来源：评论

学校读者我要写书评

暂无评论

Semantics of probabilistic Programs using s-Finite Kernels in Coq 2023

Semantics of Probabilistic Programs using s-Finite Kernels i...

引用

12th ACM SIGPLAN International Conference on Certified Programs and Proofs (CPP)

作者： Affeldt, Reynald Cohen, Cyril Saito, Ayumu Natl Inst Adv Ind Sci & Technol Tokyo Japan Univ Cote Azur Sophia Antipolis France Inria Sophia Antipolis France Tokyo Inst Technol Dept Math & Comp Sci Tokyo Japan

ISBN: (纸本)9798400700262

probabilistic programming languages are used to write probabilistic models to make probabilistic inferences. A number of rigorous semantics have recently been proposed that are now available to carry out formal verification of probabilistic programs. In this paper, we extend an existing formalization of measure and integration theory with s-finite kernels, a mathematical structure to interpret typing judgments in the semantics of a probabilistic programming language. The resulting library makes it possible to reason formally about transformations of probabilistic programs and their execution.

关键词： Coq measure theory integration theory probabilistic programming language

来源：评论

学校读者我要写书评

暂无评论

Towards verified stochastic variational inference for probabilistic programs

Towards verified stochastic variational inference for probab...

引用

Proceedings of the ACM on programming languages

作者： Lee, Wonyeol Yu, Hangyeol Rival, Xavier Yang, Hongseok School of Computing KAIST Korea Republic of INRIA Paris Département d'Informatique of ENS CNRS/PSL University France

probabilistic programming is the idea of writing models from statistics and machine learning using program notations and reasoning about these models using generic inference engines. Recently its combination with deep learning has been explored intensely, which led to the development of so called deep probabilistic programming languages, such as Pyro, Edward and ProbTorch. At the core of this development lie inference engines based on stochastic variational inference algorithms. When asked to find information about the posterior distribution of a model written in such a language, these algorithms convert this posterior-inference query into an optimisation problem and solve it approximately by a form of gradient ascent or descent. In this paper, we analyse one of the most fundamental and versatile variational inference algorithms, called score estimator or REINFORCE, using tools from denotational semantics and program analysis. We formally express what this algorithm does on models denoted by programs, and expose implicit assumptions made by the algorithm on the models. The violation of these assumptions may lead to an undefined optimisation objective or the loss of convergence guarantee of the optimisation process. We then describe rules for proving these assumptions, which can be automated by static program analyses. Some of our rules use nontrivial facts from continuous mathematics, and let us replace requirements about integrals in the assumptions, such as integrability of functions defined in terms of programs' denotations, by conditions involving differentiation or boundedness, which are much easier to prove automatically (and manually). Following our general methodology, we have developed a static program analysis for the Pyro programming language that aims at discharging the assumption about what we call model-guide support match. Our analysis is applied to the eight representative model-guide pairs from the Pyro webpage, which include sophisticated neural ne

关键词： Static analysis Computer programming languages Deep learning Engines Functions Inference engines Object oriented programming Semantics Stochastic systems Correctness Denotational semantics Posterior distributions probabilistic programming probabilistic programming language probabilistic programs Static program analysis Variational inference

来源：评论

学校读者我要写书评

暂无评论

Extending Rationality 41

Extending Rationality

引用

41st Annual Meeting of the Cognitive Science Society: Creativity + Cognition + Computation, CogSci 2019

作者： Pothos, Emmanuel M. Busemeyer, Jerome R. Pleskac, Tim Yearsley, James M. Tenenbaum, Joshua B. Goodman, Noah D. Tessler, Michael Henry Griffiths, Thomas L. Lieder, Falk Hertwig, Ralph Pachur, Thorsten Leuker, Christina Shiffrin, Richard M. Department of Psychology City University of London Northampton Square LondonEC1V 0HB United Kingdom Department of Psychological and Brain Sciences Indiana University BloomingtonIN47405 United States Department of Psychology University of Kansas LawrenceKS66045 United States Department of Brain and Cognitive Sciences Massachusetts Institute of Technology CambridgeMA02139 United States Department of Psychology Stanford University StanfordCA94305 United States Department of Psychology Princeton University PrincetonNJ08540 United States Max Planck Institute for Intelligent Systems Tübingen72076 Germany Center for Adaptive Rationality Max Planck Institute for Human Development Berlin14195 Germany

来源：评论

学校读者我要写书评

暂无评论

PγωNK: Functional probabilistic NetKAT

PγωNK: Functional probabilistic NetKAT

引用

Proceedings of the ACM on programming languages

作者： Vandenbroucke, Alexander Schrijvers, Tom KU Leuven Celestijnenlaan 200 A Leuven3001 Belgium

This work presents PγωNK, a functional probabilistic network programming language that extends probabilistic NetKAT (PNK). Like PNK, it enables probabilistic modelling of network behaviour, by providing probabilistic choice and infinite iteration (to simulate looping network packets). Yet, unlike PNK, it also offers abstraction and higher-order functions to make programming much more convenient. The formalisation of PγωNK is challenging for two reasons: Firstly, network programming induces multiple side effects (in particular, parallelism and probabilistic choice) which need to be carefully controlled in a functional setting. Our system uses an explicit syntax for thunks and sequencing which makes the interplay of these effects explicit. Secondly, measure theory, the standard domain for formalisations of (continuous) probablistic languages, does not admit higher-order functions. We address this by leveraging ω-Quasi Borel Spaces (ωQBSes), a recent advancement in the domain theory of probabilistic programming languages. We believe that our work is not only useful for bringing abstraction to PNK, but that-as part of our contribution-we have developed the meta-theory for a probabilistic language that combines advanced features like higher-order functions, iteration and parallelism, which may inform similar meta-theoretic efforts. © 2020 Copyright held by the owner/author(s).

关键词： Functional programming Abstracting Computer programming languages Higher order functions NetKAT Network modelling probabilistic language probabilistic modelling probabilistic programming probabilistic programming language Quasi Borel Spaces

来源：评论

学校读者我要写书评

暂无评论

Toward an idiomatic framework for cognitive robotics

引用

PATTERNS 2022年第7期3卷 100533页

作者： Damgaard, Malte Rormose Pedersen, Rasmus Bak, Thomas Aalborg Univ Dept Elect Syst Automat & Control DK-9220 Aalborg Denmark

Inspired by the "cognitive hourglass'' model presented by the researchers behind the cognitive architecture called Sigma, we propose a framework for developing cognitive architectures for cognitive robotics. The main purpose of the proposed framework is to ease development of cognitive architectures by encouraging cooperation and re-use of existing results. This is done by proposing a framework dividing development of cognitive architectures into a series of layers that can be considered partly in isolation, some of which directly relate to other research fields. Finally, we introduce and review some topics essential for the proposed framework. We also outline a set of applications.

关键词： robotics cognitive robotics cognitive architecture framework probabilistic programming probabilistic programming language probabilistic inference variational inference message-passing algorithm stochastic variational inference probabilistic graphical model

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：