8,530 Hits in 4.9 sec

Variational Inference for Sparse and Undirected Models [article]

John Ingraham, Debora Marks
2017 arXiv   pre-print
We find that, together, these methods for variational inference substantially improve learning of sparse undirected graphical models in simulated and real problems from physics and biology.  ...  The first is Persistent VI, an algorithm for variational inference of discrete undirected models that avoids doubly intractable MCMC and approximations of the partition function.  ...  Acknowledgements We thank David Duvenaud, Finale Doshi-Velez, Miriam Huntley, Chris Sander, and members of the Marks lab for helpful comments and discussions.  ... 
arXiv:1602.03807v2 fatcat:3s6lblnx6zajxoyh4bhmq2jaoq

Relevance Topic Model for Unstructured Social Group Activity Recognition

Fang Zhao, Yongzhen Huang, Liang Wang, Tieniu Tan
2013 Neural Information Processing Systems  
An efficient variational EM algorithm is presented for model parameter estimation and inference.  ...  In our approach, sparse Bayesian learning is incorporated into an undirected topic model (i.e., Replicated Softmax) to discover topics which are relevant to video classes and suitable for prediction.  ...  Tsinghua National Laboratory for Information Science and Technology Crossdiscipline Foundation.  ... 
dblp:conf/nips/ZhaoHWT13 fatcat:nuegfuwv7zdlpmotxwrqnmk7wy

Mouse obesity network reconstruction with a variational Bayes algorithm to employ aggressive false positive control

Benjamin A Logsdon, Gabriel E Hoffman, Jason G Mezey
2012 BMC Bioinformatics  
enter the model without the need for cross-validation or a model selection criterion.  ...  Our algorithm is the only scalable method for regularized network recovery that employs Bayesian model averaging and that can internally estimate an appropriate level of sparsity to ensure few false positives  ...  Acknowledgements We thank Larsson Omberg, Rami Mahdi, Thomas Vincent, and Jean-Luc Jannik for discussion and for their comments on this manuscript.  ... 
doi:10.1186/1471-2105-13-53 pmid:22471599 pmcid:PMC3338387 fatcat:4ebwp2drtrgczgqovh6qrwjdei

Spike-and-Slab Sparse Coding for Unsupervised Feature Discovery [article]

Ian J. Goodfellow and Aaron Courville and Yoshua Bengio
2012 arXiv   pre-print
Since exact inference in this model is intractable, we derive a structured variational inference procedure and employ a variational EM training algorithm.  ...  The S3C model resembles both the spike-and-slab RBM and sparse coding.  ...  The computation done for this work was conducted in part on computers of RESMIQ, Clumeq and SharcNet.  ... 
arXiv:1201.3382v2 fatcat:3iu2wvx3trbdnjmpd7zbfwwpmm

A Deep Latent Space Model for Graph Representation Learning [article]

Hanxuan Yang, Qingchao Kong, Wenji Mao
2021 arXiv   pre-print
For fast inference, the stochastic gradient variational Bayes (SGVB) is adopted using a non-iterative recognition model, which is much more scalable than traditional MCMC-based methods.  ...  Traditional Bayesian-based graph models and recent deep learning based GNN either suffer from impracticability or lack interpretability, thus combined models for undirected graphs have been proposed to  ...  in for undirected networks.  ... 
arXiv:2106.11721v1 fatcat:7letwm24afbb5ccxao5u5o6hzm

Graphical Modeling for High Dimensional Data

Munni Begum, Jay Bagga, C. Ann Blakey
2012 Journal of Modern Applied Statistical Methods  
A methodology based on probability and graph theory, termed graphical models, is applied to study the structure and inference of such high-dimensional data.  ...  With advances in science and information technologies, many scientific fields are able to meet the challenges of managing and analyzing high-dimensional data.  ...  This principle depends on a Consider variational inference approaches for the exponential family representations of the graphical models.  ... 
doi:10.22237/jmasm/1351743360 fatcat:eljmhzdac5cntcdybnvcvhderi

Probabilistic graphical models for climate data analysis

Arindam Banerjee
2011 Proceedings of the 2011 workshop on Climate knowledge discovery - CKD '11  
• Key Challenges -High-dimensional dependent data, small sample size -Spatial and temporal dependencies, temporal lags -Oscillations with frequency and phase variations -Important variables are unreliable  ...  variables -Example: Bayesian networks, Hidden Markov Models -Joint distribution is a product of P(child|parents) • Undirected Graphs -An undirected graph between random variables -Example: Markov  ...  Results: Droughts starting in 1960-70s The prolonged drought in Sahel in the 1970s Drought in India and Bangladesh in the 1960s Major Droughts: 1901 Droughts: -2006 18 Learning dependencies , 2002  ... 
doi:10.1145/2110230.2110235 fatcat:p3gxemm6oze7pnazykks66yngi

A linear programming approach for estimating the structure of a sparse linear genetic network from transcript profiling data

Sahely Bhadra, Chiranjib Bhattacharyya, Nagasuma R Chandra, I Saira Mian
2009 Algorithms for Molecular Biology  
The Dialogue for Reverse Engineering Assessments and Methods (DREAM) initiative provides gold standard data sets and evaluation metrics that enable and facilitate the comparison of algorithms for deducing  ...  , and the problem of estimating the structure of such a sparse linear genetic network (SLGN) from transcript profiling data.  ...  National Institute on Aging and U.S. Department of Energy (OBER). CB and NC are supported by a grant from MHRD, Government of India.  ... 
doi:10.1186/1748-7188-4-5 pmid:19239685 pmcid:PMC2654898 fatcat:jdhsnh3zyvdm3a2zyqvxmvmt6y

Variational Bayesian Methods For Multimedia Problems

Zhaofu Chen, S. Derin Babacan, Rafael Molina, Aggelos K. Katsaggelos
2014 IEEE transactions on multimedia  
In this paper we present an introduction to Variational Bayesian (VB) methods in the context of probabilistic graphical models, and discuss their application in multimedia related problems.  ...  LBP, traditionally developed using graphical models, can also be viewed as a VB inference procedure.  ...  In this paper we have provided an overview of variational Bayesian modeling and inference methods for multimedia and related areas based on the use of probabilistic graphical models.  ... 
doi:10.1109/tmm.2014.2307692 fatcat:4btc3ek37neulcmo72v6d3qlzm

Sparse Signal Recovery and Acquisition with Graphical Models

Volkan Cevher, Piotr Indyk, Lawrence Carin, Richard Baraniuk
2010 IEEE Signal Processing Magazine  
Coding this structure using an appropriate model enables JPEG2000 and other similar algorithms to compress images close to the maximum amount possible, and significantly better than a naive coder that  ...  As we will discover, GMs are not only useful for representing the prior information on x, but also lay the foundations for new kinds of measurement systems. GMs enable  ...  Relevant graphical models In preliminary work, we have identified a range of both directed and undirected graphical models that are promising for new CS recovery and inference techniques.  ... 
doi:10.1109/msp.2010.938029 fatcat:6bopdzqr4bda3dzpaof2tqnqlq

Large-Scale Feature Learning With Spike-and-Slab Sparse Coding [article]

Ian Goodfellow, Aaron Courville (Universite de Montreal), Yoshua Bengio
2012 arXiv   pre-print
We present a novel inference procedure for appropriate for use with GPUs which allows us to dramatically increase both the training set size and the amount of latent factors that S3C may be trained with  ...  In order to overcome the low amount of labeled examples available in this setting, we introduce a new feature learning and extraction procedure based on a factor model we call spike-and-slab sparse coding  ...  Our use of variational inference makes the S3C framework well-suited to integrate into the known successful strategies for learning and inference in DBM models.  ... 
arXiv:1206.6407v1 fatcat:jel6pxvwxra7phypkwzqzqaf7u

Distributed variational sparse Bayesian learning for sensor networks

Thomas Buchgraber, Dmitriy Shutin
2012 2012 IEEE International Workshop on Machine Learning for Signal Processing  
For general loopy networks, dSBL and cSBL are differend, yet simulations show much faster convergence over the variational inference iterations at similar sparsity and mean squared error performance.  ...  The proposed method is based on a combination of variational inference and loopy belief propagation, where data is only communicated between neighboring nodes without the need for a fusion center.  ...  One of the approaches for finding sparse models, which lays down the foundation for this work, is based on sparse Bayesian learning (SBL) [4, 5, 6] , exemplified by relevance vector machines (RVMs).  ... 
doi:10.1109/mlsp.2012.6349800 dblp:conf/mlsp/BuchgraberS12 fatcat:zystuqwhivhtloq7oislhqndoe

Efficient Online Inference for Bayesian Nonparametric Relational Models

Dae Il Kim, Prem Gopalan, David M. Blei, Erik B. Sudderth
2013 Neural Information Processing Systems  
Focusing on assortative models of undirected networks, we also propose an efficient structured mean field variational bound, and online methods for automatically pruning unused communities.  ...  Compared to state-of-the-art online learning methods for parametric relational models, we show significantly improved perplexity and link prediction accuracy for sparse networks with tens of thousands  ...  The infinite multiple membership relational model (IMRM) [15] also uses an IBP to allow multiple memberships, but uses a non-conjugate observation model to allow more scalable inference for sparse networks  ... 
dblp:conf/nips/KimGBS13 fatcat:psk4chejvrgbnk7kp5pstxpymi

Variational Graph Auto-Encoders [article]

Thomas N. Kipf, Max Welling
2016 arXiv   pre-print
This model makes use of latent variables and is capable of learning interpretable latent representations for undirected graphs.  ...  We introduce the variational graph auto-encoder (VGAE), a framework for unsupervised learning on graph-structured data based on the variational auto-encoder (VAE).  ...  Acknowledgments We would like to thank Christos Louizos, Mart van Baalen, Taco Cohen, Dave Herman, Pramod Sinha and Abdul-Saboor Sheikh for insightful discussions.  ... 
arXiv:1611.07308v1 fatcat:cmkw6iukwnebhcrkggzzk6iugu

Variational Bayesian Inference Techniques

Matthias Seeger, David Wipf
2010 IEEE Signal Processing Magazine  
and Bayesian graphical model technology.  ...  We describe novel variational relaxations of Bayesian integration, characterized as well as posterior maximization, which can be solved robustly for very large models by algorithms unifying convex reconstruction  ...  is a convex opti-■ mization problem if and only if MAP estimation is convex for the same model (see the section "Algorithms for Variational Sparse Bayesian Inference").  ... 
doi:10.1109/msp.2010.938082 fatcat:mntqc5cp2jfatjuglpmaal4gde
« Previous Showing results 1 — 15 out of 8,530 results