Filters








176,079 Hits in 6.3 sec

Extension of Path Probability Method to Approximate Inference over Time [article]

Vinay Jethava
2009 arXiv   pre-print
The main contribution of this thesis are the extended GBP algorithm, the extension of Path Probability Methods to the DynBP algorithm and the relationship between them.  ...  This thesis explores the Path Probability Method, a variational technique in statistical mechanics, in the context of graphical models and approximate inference problems.  ...  We shall revisit the GBP algorithm in Chapter 4 to consider the extension of the GBP algorithm to inference over time.  ... 
arXiv:0909.3606v1 fatcat:nvqd6p2t5nam5jmgq3yj6f6vgi

What we talk about when we talk about fields [article]

Ewan Cameron
2014 arXiv   pre-print
Bayesian inference offers an increasingly-popular strategy to overcome the inherent ill-posedness of this signal reconstruction challenge.  ...  In astronomical and cosmological studies one often wishes to infer some properties of an infinite-dimensional field indexed within a finite-dimensional metric space given only a finite collection of noisy  ...  of cluster-computing, the INLA method enables approximate field inference on desktop computers with run-times small enough to allow for the important follow-up inference steps of model testing and refinement  ... 
arXiv:1406.6371v1 fatcat:qmvuhgkygvg5lktt2qiizai5ey

What we talk about when we talk about fields

Ewan Cameron
2014 Proceedings of the International Astronomical Union  
Bayesian inference offers an increasingly-popular strategy to overcome the inherent ill-posedness of this signal reconstruction challenge.  ...  In astronomical and cosmological studies one often wishes to infer some properties of an infinite-dimensional field indexed within a finite-dimensional metric space given only a finite collection of noisy  ...  cluster-computing, the INLA method enables approximate field inference on desktop computers with run-times small enough to allow for the important follow-up inference steps of model testing and refinement  ... 
doi:10.1017/s1743921314010722 fatcat:cuibniz76rbefd7ysdstsgupzq

DNF Sampling for ProbLog Inference [article]

Dimitar Sht. Shterionov, Angelika Kimmig, Theofrastos Mantadelis, Gerda Janssens
2010 arXiv   pre-print
In this work we introduce a new approximation method for ProbLog inference which exploits the DNF to focus sampling.  ...  Inference in probabilistic logic languages such as ProbLog, an extension of Prolog with probabilistic facts, is often based on a reduction to a propositional formula in DNF.  ...  Approximate Inference: Program Sampling An alternative approach to inference is the use of Monte Carlo methods, that is, to use the ProbLog program to generate large numbers of random subprograms and to  ... 
arXiv:1009.3798v2 fatcat:xghkjiyk5bcxhfxlhinm7hhgca

Inference of Kinetic Ising Model on Sparse Graphs

Pan Zhang
2012 Journal of statistical physics  
Our approach gives an exact result on tree graphs and a good approximation on sparse graphs, it can be seen as an extension of Belief Propagation inference of static Ising model to kinetic Ising model.  ...  Based on dynamical cavity method, we propose an approach to the inference of kinetic Ising model, which asks to reconstruct couplings and external fields from given time-dependent output of original system  ...  Acknowledgements The author would like to thank Abolfazl Ramezanpour and Riccardo Zecchina for discussing.  ... 
doi:10.1007/s10955-012-0547-1 fatcat:37peosksynhhnksapu3rbfu3ue

Nesting Probabilistic Inference [article]

Theofrastos Mantadelis, Gerda Janssens
2011 arXiv   pre-print
When doing inference in ProbLog, a probabilistic extension of Prolog, we extend SLD resolution with some additional bookkeeping.  ...  In order to support nested probabilistic inference we propose the notion of a parametrised ProbLog engine. Nesting becomes possible by suspending and resuming instances of ProbLog engines.  ...  Acknowledgements We want to thank Paulo Moura for his advices on how to avoid nasty hacks in the implementation and the anonymous reviewers for their constructive comments.  ... 
arXiv:1112.3785v1 fatcat:343hgmmbnnhvbaiskeicaketi4

Better Intermediates Improve CTC Inference [article]

Tatsuya Komatsu, Yusuke Fujita, Jaesong Lee, Lukas Lee, Shinji Watanabe, Yusuke Kida
2022 arXiv   pre-print
uses predictions of previous inference for conditioning the next inference.  ...  This paper proposes a method for improved CTC inference with searched intermediates and multi-pass conditioning.  ...  distribution over the alignment A, the probability p(Y | X) in Eq. 1 is expressed as the sum of the probabilities of all possible alignment paths of Y , i.e., p(Y | X) = A∈B −1 (Y ) p(A | X). (6) To train  ... 
arXiv:2204.00176v1 fatcat:h6uqmupntrcdbpacdcf7ssulri

Efficient inference and learning in a large knowledge base

William Yang Wang, Kathryn Mazaitis, Ni Lao, William W. Cohen
2015 Machine Learning  
To address this bottleneck, we present a first-order probabilistic language called ProPPR in which approximate "local groundings" can be constructed in time independent of database size.  ...  Technically, ProPPR is an extension to stochastic logic programs that is biased towards short derivations; it is also closely related to an earlier relational learning algorithm called the path ranking  ...  Acknowledgments This work was sponsored in part by DARPA grant FA87501220342 to CMU and a Google Research Award. We thank Tom Mitchell and the anonymous reviewers for their helpful comments.  ... 
doi:10.1007/s10994-015-5488-x fatcat:fnm4vho5abdejiuvgsc6abtxta

Bayesian Inference for Multistate 'Step and Turn' Animal Movement in Continuous Time

A. Parton, P. G. Blackwell
2017 Journal of Agricultural Biological and Environmental Statistics  
An extension to the inference method described here allows for such a parameter to be sampled as a Gibbs step, and the path reconstruction method can be extended to include error around observed locations  ...  Although the approach for inference here is an approximation to the underlying continuous-time model, advantages remain over discrete time: behavioural switching can occur continuously in contrast to strictly  ... 
doi:10.1007/s13253-017-0286-5 fatcat:pxlk75a4m5bdbesmlrbaj7kzx4

Network Infusion to Infer Information Sources in Networks [article]

Soheil Feizi, Muriel Medard, Gerald Quon, Manolis Kellis, Ken Duffy
2016 arXiv   pre-print
Moreover, we apply NI to a real-data application, identifying news sources in the Digg social network, and demonstrate the effectiveness of NI compared to existing methods.  ...  In this paper we introduce a method called Network Infusion (NI) that has been designed to circumvent these issues, making source inference practical for large, complex real world networks.  ...  at time t if the infection reaches to it over at least one of the k disjoint shortest paths connecting that node to the source node.  ... 
arXiv:1606.07383v1 fatcat:vw2arhsiajhgxkckpwv3f5as6a

Efficient Inference and Learning in a Large Knowledge Base: Reasoning with Extracted Information using a Locally Groundable First-Order Probabilistic Logic [article]

William Yang Wang, Kathryn Mazaitis, Ni Lao, Tom Mitchell, William W. Cohen
2014 arXiv   pre-print
To address this bottleneck, we present a first-order probabilistic language called ProPPR in which that approximate "local groundings" can be constructed in time independent of database size.  ...  Technically, ProPPR is an extension to stochastic logic programs (SLPs) that is biased towards short derivations; it is also closely related to an earlier relational learning algorithm called the path  ...  Acknowledgements This work was sponsored in part by DARPA grant FA87501220342 to CMU and a Google Research Award.  ... 
arXiv:1404.3301v1 fatcat:s5q4g22bybgkhnghdm75jhzjwy

Dynamic Obstacle Avoidance Using Bayesian Occupancy Filter and Approximate Inference

Ángel Llamazares, Vladimir Ivan, Eduardo Molinos, Manuel Ocaña, Sethu Vijayakumar
2013 Sensors  
While several obstacle avoidance systems have been presented in the literature addressing safety and optimality of the robot motion separately, we have applied the approximate inference framework to this  ...  The goal of this paper is to solve the problem of dynamic obstacle avoidance for a mobile platform using the stochastic optimal control framework to compute paths that are optimal in terms of safety and  ...  Acknowledgements This work has been funded by TIN2011-29824-C02-01 and TIN2011-29824-C02-02 (ABSYNTHE project) from the "Ministerio de Economía y Competitividad" and a grant of the Fundación Caja Madrid  ... 
doi:10.3390/s130302929 pmid:23529117 pmcid:PMC3658723 fatcat:j35go3nbyrd45ejqvhrgzwbpg4

Variational Program Inference [article]

Georges Harik, Noam Shazeer
2010 arXiv   pre-print
We introduce a framework for representing a variety of interesting problems as inference over the execution of probabilistic model programs.  ...  In addition, we show how to use the guide program as a proposal distribution in importance sampling to statistically prove lower bounds on the probability of the evidence and on the probability of a hypothesis  ...  Acknowledgments Thanks to Jeremy Bem for introducing us to program induction, Sergey Pankov to importance sampling, and Noah Goodman, Vikash Mansinghka, Daniel Roy for a discussion on extra choices.  ... 
arXiv:1006.0991v1 fatcat:7zesldiunbbz7eyrfki6m6xcpi

On the implementation of the probabilistic logic programming language ProbLog

ANGELIKA KIMMIG, BART DEMOEN, LUC DE RAEDT, VÍTOR SANTOS COSTA, RICARDO ROCHA
2011 Theory and Practice of Logic Programming  
ProbLog is a recent probabilistic extension of Prolog motivated by the mining of large biological networks. In ProbLog, facts can be labeled with probabilities.  ...  These facts are treated as mutually independent random variables that indicate whether these facts belong to a randomly sampled program. Different kinds of queries can be posed to ProbLog programs.  ...  the development of SimpleCUDD.  ... 
doi:10.1017/s1471068410000566 fatcat:q3nliq4bpnca3ohmwsz2ywjaty

Bayesian Verification of Chemical Reaction Networks [article]

Gareth W. Molyneux, Viraj B. Wijesuriya, Alessandro Abate
2020 arXiv   pre-print
Secondly, we utilise Bayesian inference to update a probability distribution of the parameters within a parametric model with data gathered from the underlying CRN.  ...  In the third and final stage, we combine the results of both steps to compute the probability that the underlying CRN satisfies the given property.  ...  When inferring parameters, we compute a probability distribution over the set of inferred parameters.  ... 
arXiv:2004.11321v1 fatcat:rmcmmv2lvrf65mdyovqcfm4czy
« Previous Showing results 1 — 15 out of 176,079 results