We present AQUA, a new probabilistic inference algorithm that operates on probabilistic programs with continuous posterior distributions. AQUA approximates programs via an efficient quantization of the continuous dist...
详细信息
We present AQUA, a new probabilistic inference algorithm that operates on probabilistic programs with continuous posterior distributions. AQUA approximates programs via an efficient quantization of the continuous distributions. It represents the distributions of random variables using quantized value intervals (Interval Cube) and corresponding probability densities (Density Cube). AQUA's analysis transforms Interval and Density Cubes to compute the posterior distribution with bounded error. We also present an adaptive algorithm for selecting the size and the granularity of the Interval and Density Cubes. We evaluate AQUA on 24 programs from the literature. AQUA solved all of 24 benchmarks in less than 43s (median 1.35s) with a high level of accuracy. We show that AQUA is more accurate than state-of-the-art approximate algorithms (Stan's NUTS and ADVI) and supports programs that are out of reach of exact inference tools, such as PSI and SPPL.
We present a major new version of Scenic, a probabilistic programming language for writing formal models of the environments of cyber-physical systems. Scenic has been successfully used for the design and analysis of ...
详细信息
ISBN:
(数字)9783031377068
ISBN:
(纸本)9783031377051;9783031377068
We present a major new version of Scenic, a probabilistic programming language for writing formal models of the environments of cyber-physical systems. Scenic has been successfully used for the design and analysis of CPS in a variety of domains, but earlier versions are limited to environments that are essentially two-dimensional. In this paper, we extend Scenic with native support for 3D geometry, introducing new syntax that provides expressive ways to describe 3D configurations while preserving the simplicity and readability of the language. We replace Scenic's simplistic representation of objects as boxes with precise modeling of complex shapes, including a ray tracing-based visibility system that accounts for object occlusion. We also extend the language to support arbitrary temporal requirements expressed in LTL, and build an extensible Scenic parser generated from a formal grammar of the language. Finally, we illustrate the new application domains these features enable with case studies that would have been impossible to accurately model in Scenic 2.
probabilistic programming languages (PPLs) are essential for reasoning under uncertainty. Even though many real-world probabilistic programs involve discrete distributions, the state-of-the-art PPLs are suboptimal for...
详细信息
ISBN:
(纸本)9798400700880
probabilistic programming languages (PPLs) are essential for reasoning under uncertainty. Even though many real-world probabilistic programs involve discrete distributions, the state-of-the-art PPLs are suboptimal for a large class of tasks dealing with such distributions. In this paper, we propose BayesTensor, a tensor-based probabilistic programming framework. By generating tensor algebra code from probabilistic programs, BayesTensor takes advantage of the highly-tuned vectorized implementations of tensor processing frameworks. Our experiments show that BayesTensor outperforms the state-of-the-art frameworks in a variety of discrete probabilistic programs, inference over Bayesian Networks, and real-world probabilistic programs employed in data processing systems.
Much work has been done to give semantics to probabilistic programming languages. In recent years, most of the semantics used to reason about probabilistic programs fall in two categories: semantics based on Markov ke...
详细信息
ISBN:
(数字)9783031308291
ISBN:
(纸本)9783031308284;9783031308291
Much work has been done to give semantics to probabilistic programming languages. In recent years, most of the semantics used to reason about probabilistic programs fall in two categories: semantics based on Markov kernels and semantics based on linear operators. Both styles of semantics have found numerous applications in reasoning about probabilistic programs, but they each have their strengths and weaknesses. Though it is believed that there is a connection between them there are no languages that can handle both styles of programming. In this work we address these questions by defining a two-level calculus and its categorical semantics which makes it possible to program with both kinds of semantics. From the logical side of things we see this language as an alternative resource interpretation of linear logic, where the resource being kept track of is sampling instead of variable use.
Inversion is a fundamental operation that arises frequently in probabilistic inference and computer graphics. For example, inversion is used to decrease variance and to enable differentiation in variational inference ...
详细信息
ISBN:
(纸本)9798400712142
Inversion is a fundamental operation that arises frequently in probabilistic inference and computer graphics. For example, inversion is used to decrease variance and to enable differentiation in variational inference (e.g., reparameterization trick) and in differentiable rendering (e.g., to integrate over object boundaries). Existing approaches to inversion limit the class of functions inverted, for example, to affine functions, or require a user-specified inverse. We study when a local inverse-an inverse that is valid in a neighborhood of a point-exists. We provide an algorithm to approximate the local inverse and give the convergence rate of the solver. We present LIN, a system that automatically computes the local inverse of a function using a fixed-point solver. We implement LIN in Python and use it to automatically compute the local inverse of affine, polar, and hyperbolic changes of variables arising in image stylization.
Extending programming languages with stochastic behaviour such as probabilistic choices or random sampling has a long tradition in computer science. A recent development in this direction is a declarative probabilisti...
详细信息
ISBN:
(纸本)9798400701276
Extending programming languages with stochastic behaviour such as probabilistic choices or random sampling has a long tradition in computer science. A recent development in this direction is a declarative probabilistic programming language, proposed by Barany et al. in 2017, which operates on standard relational databases. In particular, Barany et al. proposed generative Datalog, a probabilistic extension of Datalog that allows sampling from discrete probability distributions. Intuitively, the output of a generative Datalog program Pi on an input database D is a probability space over the minimal models of D and Pi, the so-called possible outcomes. This is a natural generalization of the (deterministic) semantics of Datalog, where the output of a program on a database is their unique minimal model. A natural question to ask is how generative Datalog can be enriched with the useful feature of negation, which in turn leads to a strictly more expressive declarative probabilistic programming language. In particular, the challenging question is how the probabilistic semantics of generative Datalog with negation can be robustly defined. Our goal is to provide an answer to this question by interpreting negation according to the stable model semantics.
probabilistic answer set programming (PASP) combines rules, facts, and independent probabilistic facts. Often one restricts such programs so that every query yields a sharp probability value. The purpose of this paper...
详细信息
probabilistic answer set programming (PASP) combines rules, facts, and independent probabilistic facts. Often one restricts such programs so that every query yields a sharp probability value. The purpose of this paper is to argue that a very useful modeling language is obtained by adopting a particular credal semantics for PASP, where one associates with each consistent program a credal set. We examine the basic properties of PASP and present an algorithm to compute (upper) probabilities given a program.
We consider a programming language that can manipulate both classical and quantum information. Our language is type-safe and designed for variational quantum programming, which is a hybrid classical-quantum computatio...
详细信息
We consider a programming language that can manipulate both classical and quantum information. Our language is type-safe and designed for variational quantum programming, which is a hybrid classical-quantum computational paradigm. The classical subsystem of the language is the probabilistic FixPoint Calculus (PFPC), which is a lambda calculus with mixed-variance recursive types, term recursion and probabilistic choice. The quantum subsystem is a first-order linear type system that can manipulate quantum information. The two subsystems are related by mixed classical/quantum terms that specify how classical probabilistic effects are induced by quantum measurements, and conversely, how classical (probabilistic) programs can influence the quantum dynamics. We also describe a sound and computationally adequate denotational semantics for the language. Classical probabilistic effects are interpreted using a recently-described commutative probabilistic monad on DCPO. Quantum effects and resources are interpreted in a category of von Neumann algebras that we show is enriched over (continuous) domains. This strong sense of enrichment allows us to develop novel semantic methods that we use to interpret the relationship between the quantum and classical probabilistic effects. By doing so we provide a very detailed denotational analysis that relates domain-theoretic models of classical probabilistic programming to models of quantum programming.
We introduce a novel rule-based approach for handling regression problems. The new methodology carries elements from two frameworks: (i) it provides information about the uncertainty of the parameters of interest usin...
详细信息
We introduce a novel rule-based approach for handling regression problems. The new methodology carries elements from two frameworks: (i) it provides information about the uncertainty of the parameters of interest using Bayesian inference, and (ii) it allows the incorporation of expert knowledge through rule-based systems. The blending of those two different frameworks can be particularly beneficial for various domains (e.g., engineering), where even though the significance of uncertainty quantification motivates a Bayesian approach, there is no simple way to incorporate researcher intuition into the model. We validate our models by applying them to synthetic applications: a simple linear regression problem and two more complex structures based on partial differential equations, and we illustrate their use through two cases derived from real data. Finally, we review the advantages of our methodology, which include the simplicity of the implementation, the uncertainty reduction due to the added information and, in some occasions, the derivation of better point predictions, and we outline limitations, mainly from the computational complexity perspective, such as the difficulty in choosing an appropriate algorithm and the added computational burden.
probabilistic programming languages (PPLs) allow programmers to construct statistical models and then simulate data or perform inference over them. Many PPLs restrict models to a particular instance of simulation or i...
详细信息
probabilistic programming languages (PPLs) allow programmers to construct statistical models and then simulate data or perform inference over them. Many PPLs restrict models to a particular instance of simulation or inference, limiting their reusability. In other PPLs, models are not readily composable. Using Haskell as the host language, we present an embedded domain specific language based on algebraic effects, where probabilistic models are modular, first-class, and reusable for both simulation and inference. We also demonstrate how simulation and inference can be expressed naturally as composable program transformations using algebraic effect handlers.
暂无评论