Filters








3,489 Hits in 3.0 sec

Approximate lifted inference with probabilistic databases

Wolfgang Gatterbauer, Dan Suciu
2015 Proceedings of the VLDB Endowment  
This paper proposes a new approach for approximate evaluation of #P-hard queries with probabilistic databases.  ...  We give a detailed experimental evaluation of our approach and, in the process, provide a new way of thinking about the value of probabilistic methods over non-probabilistic methods for ranking query answers  ...  Lifted and approximate inference.  ... 
doi:10.14778/2735479.2735494 fatcat:safh6aiiwfdlxpitp3bipdxg7y

Approximate Lifted Inference with Probabilistic Databases [article]

Wolfgang Gatterbauer, Dan Suciu
2014 arXiv   pre-print
This paper proposes a new approach for approximate evaluation of #P-hard queries with probabilistic databases.  ...  We give a detailed experimental evaluation of our approach and, in the process, provide a new way of thinking about the value of probabilistic methods over non-probabilistic methods for ranking query answers  ...  Lifted and approximate inference.  ... 
arXiv:1412.1069v1 fatcat:pqzm45dp5rh6njv5wrybpdj3xq

10 Years of Probabilistic Querying – What Next? [chapter]

Martin Theobald, Luc De Raedt, Maximilian Dylla, Angelika Kimmig, Iris Miliaraki
2013 Lecture Notes in Computer Science  
techniques based on knowledge compilation and lifted (first-order) inference.  ...  Over the past decade, the two research areas of probabilistic databases and probabilistic programming have intensively studied the problem of making structured probabilistic inference scalable, but-so  ...  Probabilistic programming provides sophisticated knowledge compilation techniques and initial approaches for lifted (first-order) inference, with judiciously tuned approximation algorithms for the cases  ... 
doi:10.1007/978-3-642-40683-6_1 fatcat:lofuquzqgbb4hcjtjeqydyakbe

Query Processing over Uncertain Data [chapter]

Nilesh Dalvi, Dan Olteanu
2017 Encyclopedia of Database Systems  
In the probabilistic setting, this problem becomes the computation of all pairs (t, p), where the tuple t is in the query answer in some random world of the input probabilistic database with probability  ...  SYNONYMS Query Processing over Probabilistic Data DEFINITION An uncertain or probabilistic database is defined as a probability distribution over a set of deterministic database instances called possible  ...  SPROUT pioneered in-and out-database techniques based on lin-eage compilation [11] . Reminiscent of lifted inference in AI, follow-up work lifts compilation to first-order lineage [9] .  ... 
doi:10.1007/978-1-4899-7993-3_80689-1 fatcat:fesows5udzcgpcihitkwzlpeze

Scaling Lifted Probabilistic Inference and Learning Via Graph Databases

Mayukh Das, Yuqing Wu, Tushar Khot, Kristian Kersting, Sriraam Natarajan
2016 Proceedings of the 2016 SIAM International Conference on Data Mining  
This paper investigates whether 'Compilation to Graph Databases' could be a practical technique for scaling lifted probabilistic inference and learning methods.  ...  One of the key operations inside these lifted approaches is counting -be it for parameter/structure learning or for efficient inference.  ...  Lifted Inference As a final task, given the recent surge in interest in the so-called lifted probabilistic inference methods, we employed our approximation strategy for counting in one such method.  ... 
doi:10.1137/1.9781611974348.83 dblp:conf/sdm/DasWKKN16 fatcat:6ng7pbllvva2vh3b2fcuqfeaaa

State-Space Abstractions for Probabilistic Inference: A Systematic Review [article]

Stefan Lüdtke, Max Schröder, Frank Krüger, Sebastian Bader, Thomas Kirste
2018 arXiv   pre-print
Algorithms that exploit those symmetries have been devised in different research fields, for example by the lifted inference-, multiple object tracking-, and modeling and simulation-communities.  ...  Researchers from different fields that are confronted with a state space explosion problem in a probabilistic system can use this classification to identify possible solutions.  ...  Several other search- Probabilistic Databases Ideas related to lifted inference arose independently in the probabilistic database community.  ... 
arXiv:1804.06748v2 fatcat:yj2eafizonaqvcg7k4e24tacya

Query Processing on Probabilistic Data: A Survey

Guy Van den Broeck, Dan Suciu
2017 Foundations and Trends in Databases  
SlimShot SlimShot was introduced by Gribkoff and Suciu [2016] as a probabilistic database system that combines lifted inference with approximate weighted model counting based on Monte Carlo simulations  ...  On the theoretical side, one is to study the complexity of approximate lifted inference.  ...  Recall that the database schema is R = (R 1 , R 2 , . . . , R ). Consider the set S = {R 1 , . . . , R } ∪ [n] with the total order R 1 < R 2 < · · · < R < 1 < 2 < · · · < n.  ... 
doi:10.1561/1900000052 fatcat:jzifdhyvsnh7thqrnuptxbpejy

Bisimulation-based Approximate Lifted Inference [article]

Prithviraj Sen, Amol Deshpande, Lise Getoor
2012 arXiv   pre-print
Here, we describe lifted inference algorithms that determine symmetries and automatically lift the probabilistic model to speedup inference.  ...  We report experiments on both synthetic and real data to show that in the presence of symmetries, run-times for inference can be improved significantly, with approximate lifted inference providing orders  ...  We would also like to thank the anonymous reviewers for their comments and suggestions and Parag Singla for sharing with us the Cora-ER MLN.  ... 
arXiv:1205.2616v1 fatcat:xxzrbs4idvhkpe5p523gbfvdx4

SlimShot

Eric Gribkoff, Dan Suciu
2016 Proceedings of the VLDB Endowment  
SlimShot converts the MLN to a tuple-independent probabilistic database, then uses a simple Monte Carlo-based inference, with three key enhancements: (1) it combines sampling with safe query evaluation  ...  In this paper we describe SlimShot (Scalable Lifted Inference and Monte Carlo Sampling Hybrid Optimization Technique), a probabilistic inference engine for knowledge bases.  ...  (8) , which consists of combining sampling with lifted probabilistic inference.  ... 
doi:10.14778/2904483.2904487 fatcat:6d3qzzzeczcgnnip2uhormrcka

Evaluating Inference Algorithms for the Prolog Factor Language [chapter]

Tiago Gomes, Vítor Santos Costa
2013 Lecture Notes in Computer Science  
PFL is also capable of solving probabilistic queries on these models through the implementation of four inference algorithms: variable elimination, belief propagation, lifted variable elimination and lifted  ...  Over the last years there has been some interest in models that combine first-order logic and probabilistic graphical models to describe large scale domains, and in efficient ways to perform inference  ...  First, we would like to use the PFL to understand the general usefulness of lifted inference and how it plays with logical and probabilistic inference.  ... 
doi:10.1007/978-3-642-38812-5_6 fatcat:zmarg5erd5de7owelpw3pkyv54

Lifted generative learning of Markov logic networks

Jan Van Haaren, Guy Van den Broeck, Wannes Meert, Jesse Davis
2015 Machine Learning  
This in turn requires performing probabilistic inference, which, in general, is intractable in MLNs. Lifted inference speeds up probabilistic inference by exploiting symmetries in a model.  ...  First, we provide a generic algorithm for learning maximum likelihood weights that works with any exact lifted inference approach.  ...  speed up probabilistic inference.  ... 
doi:10.1007/s10994-015-5532-x fatcat:ifzxvwghxfbyndujkqmarwmpw4

Learning, Logic, and Probability: A Unified View [chapter]

Pedro Domingos
2004 Lecture Notes in Computer Science  
Overview  Motivation  Background  Markov logic  Inference  Learning  Software  Applications  Discussion Logical and Statistical AI Field Logical approach Statistical approach Knowledge representation  ...  graphical models and first-order logic are special cases  Unified inference algorithms  Unified learning algorithms  Easy-to-use software  Broad applicability  A new kind of programming language  ...   Can KBMC, lazy and lifted inference be combined?  Can we have lifted inference over both probabilistic and deterministic dependencies? (Lifted MC-SAT?)  ... 
doi:10.1007/978-3-540-30109-7_26 fatcat:bcnvefkpizhx5o2ojubolrz6ca

Learning, Logic, and Probability: A Unified View [chapter]

Pedro Domingos
2006 Lecture Notes in Computer Science  
Overview  Motivation  Background  Markov logic  Inference  Learning  Software  Applications  Discussion Logical and Statistical AI Field Logical approach Statistical approach Knowledge representation  ...  graphical models and first-order logic are special cases  Unified inference algorithms  Unified learning algorithms  Easy-to-use software  Broad applicability  A new kind of programming language  ...   Can KBMC, lazy and lifted inference be combined?  Can we have lifted inference over both probabilistic and deterministic dependencies? (Lifted MC-SAT?)  ... 
doi:10.1007/11874850_2 fatcat:pwtbgmnberfv5apufkkvygpbu4

Learning, Logic, and Probability: A Unified View [chapter]

Pedro Domingos
2006 Lecture Notes in Computer Science  
Overview  Motivation  Background  Markov logic  Inference  Learning  Software  Applications  Discussion Logical and Statistical AI Field Logical approach Statistical approach Knowledge representation  ...  graphical models and first-order logic are special cases  Unified inference algorithms  Unified learning algorithms  Easy-to-use software  Broad applicability  A new kind of programming language  ...   Can KBMC, lazy and lifted inference be combined?  Can we have lifted inference over both probabilistic and deterministic dependencies? (Lifted MC-SAT?)  ... 
doi:10.1007/978-3-540-36668-3_1 fatcat:7xyuyldeyrfjdhsriy5rosc3re

Learning, Logic, and Probability: A Unified View [chapter]

Pedro Domingos
2006 Lecture Notes in Computer Science  
Overview  Motivation  Background  Markov logic  Inference  Learning  Software  Applications  Discussion Logical and Statistical AI Field Logical approach Statistical approach Knowledge representation  ...  graphical models and first-order logic are special cases  Unified inference algorithms  Unified learning algorithms  Easy-to-use software  Broad applicability  A new kind of programming language  ...   Can KBMC, lazy and lifted inference be combined?  Can we have lifted inference over both probabilistic and deterministic dependencies? (Lifted MC-SAT?)  ... 
doi:10.1007/11891451_2 fatcat:3lsn5lklcrbavbn2wxo7l7hpfq
« Previous Showing results 1 — 15 out of 3,489 results