4,929 Hits in 4.5 sec

Amortized Bethe Free Energy Minimization for Learning MRFs [article]

Sam Wiseman, Yoon Kim
2019 arXiv   pre-print
We furthermore amortize this optimization with trained inference networks.  ...  We propose to learn deep undirected graphical models (i.e., MRFs) with a non-ELBO objective for which we can calculate exact gradients.  ...  In particular, we will consider the 3rd order undirected product-of-experts style HMM in Figure 1 (b) , which contains only pairwise factors, and parameterizes the joint distribution of x and z as P (  ... 
arXiv:1906.06399v2 fatcat:7mpdcioj5zavfdbe7fgdikgh2i

A simple generative model of the mouse mesoscale connectome

Sid Henriksen, Rich Pang, Mark Wronkiewicz
2016 eLife  
Recent technological advances now allow for the collection of vast data sets detailing the intricate neural connectivity patterns of various organisms.  ...  Oh et al. (2014) recently published the most complete description of the mouse mesoscale connectome acquired to date.  ...  We also thank the directors of this course, Christof Koch and Adrienne Fairhall.  ... 
doi:10.7554/elife.12366 pmid:26978793 pmcid:PMC4807721 fatcat:3pzqjvzibffrpp6pogxufl2ahi

Marginal Pseudo-Likelihood Learning of Discrete Markov Network Structures

Johan Pensar, Henrik Nyman, Juha Niiranen, Jukka Corander
2017 Bayesian Analysis  
The core of the Markov network representation is an undirected graph which elegantly captures the dependence structure over the variables.  ...  Markov networks are a popular tool for modeling multivariate distributions over a set of discrete variables.  ...  HN was supported by the Foundation ofÅbo Akademi University, as part of the grant for the Center of Excellence in Optimization and Systems Engineering.  ... 
doi:10.1214/16-ba1032 fatcat:pqimkuujkvff3nst7jdv2cudym

Identification of sparse communication graphs in consensus networks

Fu Lin, Makan Fardad, Mihailo R. Jovanovic
2012 2012 50th Annual Allerton Conference on Communication, Control, and Computing (Allerton)  
We consider the design of distributed controller architectures for undirected networks of single-integrators.  ...  In the presence of stochastic disturbances, we identify communication topologies that balance the variance amplification of the network with the number of communication links.  ...  By solving a parameterized family of SDPs, we obtain a tradeoff curve between the variance amplification of the consensus network and the number of communication links in the distributed controllers.  ... 
doi:10.1109/allerton.2012.6483203 dblp:conf/allerton/LinFJ12 fatcat:zr7s75ir25barcs4bzovcnkupq

Variational Inference for Sparse and Undirected Models [article]

John Ingraham, Debora Marks
2017 arXiv   pre-print
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.  ...  parameterizations.  ...  Portions of this work were conducted on the Orchestra HPC Cluster at Harvard Medical School.  ... 
arXiv:1602.03807v2 fatcat:3s6lblnx6zajxoyh4bhmq2jaoq

Decentralized detection with long-distance communication

O. Patrick Kreidl, Alan S. Willsky
2008 2008 42nd Asilomar Conference on Signals, Systems and Computers  
We consider the well-studied decentralized Bayesian detection problem with the twist that a small subset of nodes (all arranged in a given directed network topology) may also communicate to "long-distance  ...  The fact that the leader network can connect nodes that are not spatial neighbors in the non-leader network captures the opportunity for "long-distance" communication.  ...  The (b) "leader" network is defined by any undirected graph G U , connecting via "long-distance" links an arbitrary yet relatively small subset of the nodes in (a).  ... 
doi:10.1109/acssc.2008.5074639 fatcat:x3rbjgqgljgqffqebqgbb77ssy

I/O-efficient Hierarchical Diameter Approximation [chapter]

Deepak Ajwani, Ulrich Meyer, David Veith
2012 Lecture Notes in Computer Science  
Acknowledgements The authors would like to thank Andreas Beckmann for his help with STXXL and the hardware resources and the anonymous reviewers for their feedback.  ...  For connected undirected unweighted graphs G(V, E) where n = |V | and m = |E|, the distance d(u, v) between two nodes u, v ∈ V is the number of edges in the shortest path connecting u and v.  ...  Parameterized approach. The parameterized approach decomposes the input undirected graph into many small-diameter clusters and then contracts each cluster to a single node forming a condensed graph.  ... 
doi:10.1007/978-3-642-33090-2_8 fatcat:bocwwsbfarhezof2aykqhroyyu

Graphical Models in a Nutshell [chapter]

2007 Introduction to Statistical Relational Learning  
in large networks.  ...  Graphical models have enjoyed a surge of interest in the last two decades, due both to the flexibility and power of the representation and to the increased ability to effectively learn and perform inference  ...  The remaining question is how to parameterize this undirected graph. The graph structure represents the qualitative properties of the distribution.  ... 
doi:10.7551/mitpress/7432.003.0004 fatcat:wbhjah7qczdftaiod5jg4ok2xe

Causal inference, social networks, and chain graphs [article]

Elizabeth L. Ogburn, Ilya Shpitser, Youjin Lee
2020 arXiv   pre-print
We argue that, in some settings, chain graph models approximate the marginal distribution of a snapshot of a longitudinal data generating process on interacting units.  ...  Our parameterization corresponds to a particular family of graphical models known as chain graphs.  ...  In this paper we illustrate a parsimonious parameterization for such social network data and explore when this new parameterization might be justified.  ... 
arXiv:1812.04990v2 fatcat:mrovc2h2tjabzpabgd5qsp7l24

Image Disguise based on Generative Model [article]

Xintao Duan, Haoxian Song, En Zhang, Jingjing Liu
2018 arXiv   pre-print
This image disguise method not only solves the problem of obvious visual implication, but also guarantees the security of the information.  ...  well-trained generative model to achieve the effect of disguising the original image.  ...  ACKNOWLEDGMENT This work was supported by the National Natural Science Foundation of China under Grant NO.U1204606, U1404603, the Science and Technology Foundation of Henan Province under Grant No.172102210335  ... 
arXiv:1710.07782v4 fatcat:trqc7fxeirb63gedr3gjxlqyny

Percolation on Isotropically Directed Lattice [article]

Aurelio W. T. de Noronha, André A. Moreira, André P. Vieira, Hans J. Herrmann, José S. Andrade, Humberto A. Carmona
2018 arXiv   pre-print
with the same probability.  ...  We derive exact results for the percolation threshold on planar lattices, and present a conjecture for the value the percolation-threshold for in any lattice.  ...  ACKNOWLEDGMENTS The thank the Brazilian agencies CNPq, CAPES, FUNCAP, NAP-FCx, the National Institute of Science and Technology for Complex Fluids (INCT-FCx) and the National Institute of Science and Technology  ... 
arXiv:1808.06644v1 fatcat:5a6lnkuabvbdjenmuj3fccp36e

Distributed parameterized model predictive control of networked multi-agent systems

Greg Droge, Magnus Egerstedt
2013 2013 American Control Conference  
in the network.  ...  It enables both the simulation of neighbors' states as well as the ability to minimize a collective cost in a distributed fashion.  ...  In section IV, information requirements will be discussed to determine if a given network of communication is sufficient to perform distributed parameterized MPC.  ... 
doi:10.1109/acc.2013.6580021 fatcat:qkvsvjbzjvh3bob46scz3jgxg4

The Neighborhood Broadcast Problem in Wireless Ad Hoc Sensor Networks

Stefan Hoffmann, Egon Wanke
2017 International Journal on Applications of Graph Theory In wireless Ad Hoc Networks And sensor Networks  
It is parameterized with a positive integer k ∈ N and it is proven to guarantee delivery for k ≥ 2d − 1, if node v is d-locally connected, which means that the set of nodes with distance between 1 and  ...  This paper considers the following Neighborhood Broadcast problem: Distribute a message to all neighbors of a network node v under the assumption that v does not participate due to being corrupted or damaged  ...  The diameter of an undirected graph G is the maximum distance between any two vertices of G. Definition 1. Let d ∈ N be a positive integer, G = (V, E) an undirected graph and v ∈ V a vertex.  ... 
doi:10.5121/jgraphoc.2017.9101 fatcat:nbwaxyzpbvetlavitw2ckwutjy

Page 4882 of Mathematical Reviews Vol. , Issue 90H [page]

1990 Mathematical Reviews  
Summary: “The k most vital arcs in a network are those whose removal from the network results in the greatest increase in the shortest distance between two specified nodes.  ...  A maximin distribution of flows under which the relative amount of the smallest of the transmitted flows is maximum is adopted as the solution of the problem.” 90h:90071 90B10 90C35 Malik, K. (1-CRNL-GM  ... 

Identifiability of species network topologies from genomic sequences using the logDet distance [article]

Elizabeth S. Allman, Hector Baños, John A. Rhodes
2021 arXiv   pre-print
The heavy computational demands of standard approaches to this problem severely limit the size of datasets that may be analyzed, in both the number of species and the number of genetic loci.  ...  Inference of network-like evolutionary relationships between species from genomic data must address the interwoven signals from both gene flow and incomplete lineage sorting.  ...  The undirected LSA network N is the rooted network obtained from the LSA network N ⊕ by undirecting all edges. 3.  ... 
arXiv:2108.01765v1 fatcat:kh4ngfygfba73pxvjodjumxg5q
« Previous Showing results 1 — 15 out of 4,929 results