Declarative Probabilistic Programming with Datalog
نویسندگان
چکیده
منابع مشابه
PPDL: Probabilistic Programming with Datalog
There has been a substantial recent focus on the concept of probabilistic programming [6] towards its positioning as a prominent paradigm for advancing and facilitating the development of machine-learning applications. A probabilisticprogramming language typically consists of two components: a specification of a stochastic process (the prior), and a specification of observations that restrict t...
متن کاملDeclarative Statistical Modeling with Datalog
Formalisms for specifying general statistical models, such as probabilistic-programming languages, typically consist of two components: a specification of a stochastic process (the prior), and a specification of observations that restrict the probability space to a conditional subspace (the posterior). Use cases of such formalisms include the development of algorithms in machine learning and ar...
متن کاملInformation Retrieval with Probabilistic Datalog
The probabilistic logical approach in Information Retrieval (IR) aims at describing the retrieval process as the computation of the probability P (d! q) that a document d implies a query q. Probabilistic Datalog (DatalogP ) is a logic that enables uncertain inference. We use DatalogP as a platform for investigating the probabilistic logical approach in IR. The expressiveness of DatalogP allows ...
متن کاملJudgeD: A Probabilistic Datalog with Dependencies
We present JudgeD, a probabilistic datalog. A JudgeD program defines a distribution over a set of traditional datalog programs by attaching logical sentences to clauses to implicitly specify traditional data programs. Through the logical sentences, JudgeD provides a novel method for the expression of complex dependencies between both rules and facts. JudgeD is implemented as a proof-of-concept ...
متن کاملExtending Probabilistic Datalog
In [Fuhr 95], we have proposed probabilistic Datalog (pD) as an inference engine for IR. However, a closer look at pD shows that it handles conditional probabilities only in a rather limited way, since it neither allows for the speci cation of conditional probabilities in rules nor is it able to compute conditional probabilities. Given that probabilistic IR should be interpreted as computing th...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: ACM Transactions on Database Systems
سال: 2017
ISSN: 0362-5915,1557-4644
DOI: 10.1145/3132700