Publish in OALib Journal

ISSN: 2333-9721

APC: Only $99


Any time

2015 ( 4 )

2014 ( 10 )

2013 ( 10 )

2012 ( 23 )

Custom range...

Search Results: 1 - 10 of 112 matches for " Willie Neiswanger "
All listed articles are free for downloading (OA Articles)
Page 1 /112
Display every page Item
Unsupervised Detection and Tracking of Arbitrary Objects with Dependent Dirichlet Process Mixtures
Willie Neiswanger,Frank Wood
Computer Science , 2012,
Abstract: This paper proposes a technique for the unsupervised detection and tracking of arbitrary objects in videos. It is intended to reduce the need for detection and localization methods tailored to specific object types and serve as a general framework applicable to videos with varied objects, backgrounds, and image qualities. The technique uses a dependent Dirichlet process mixture (DDPM) known as the Generalized Polya Urn (GPUDDPM) to model image pixel data that can be easily and efficiently extracted from the regions in a video that represent objects. This paper describes a specific implementation of the model using spatial and color pixel data extracted via frame differencing and gives two algorithms for performing inference in the model to accomplish detection and tracking. This technique is demonstrated on multiple synthetic and benchmark video datasets that illustrate its ability to, without modification, detect and track objects with diverse physical characteristics moving over non-uniform backgrounds and through occlusion.
Asymptotically Exact, Embarrassingly Parallel MCMC
Willie Neiswanger,Chong Wang,Eric Xing
Computer Science , 2013,
Abstract: Communication costs, resulting from synchronization requirements during learning, can greatly slow down many parallel machine learning algorithms. In this paper, we present a parallel Markov chain Monte Carlo (MCMC) algorithm in which subsets of data are processed independently, with very little communication. First, we arbitrarily partition data onto multiple machines. Then, on each machine, any classical MCMC method (e.g., Gibbs sampling) may be used to draw samples from a posterior distribution given the data subset. Finally, the samples from each machine are combined to form samples from the full posterior. This embarrassingly parallel algorithm allows each machine to act independently on a subset of the data (without communication) until the final combination stage. We prove that our algorithm generates asymptotically exact samples and empirically demonstrate its ability to parallelize burn-in and sampling in several models.
Embarrassingly Parallel Variational Inference in Nonconjugate Models
Willie Neiswanger,Chong Wang,Eric Xing
Computer Science , 2015,
Abstract: We develop a parallel variational inference (VI) procedure for use in data-distributed settings, where each machine only has access to a subset of data and runs VI independently, without communicating with other machines. This type of "embarrassingly parallel" procedure has recently been developed for MCMC inference algorithms; however, in many cases it is not possible to directly extend this procedure to VI methods without requiring certain restrictive exponential family conditions on the form of the model. Furthermore, most existing (nonparallel) VI methods are restricted to use on conditionally conjugate models, which limits their applicability. To combat these issues, we make use of the recently proposed nonparametric VI to facilitate an embarrassingly parallel VI procedure that can be applied to a wider scope of models, including to nonconjugate models. We derive our embarrassingly parallel VI algorithm, analyze our method theoretically, and demonstrate our method empirically on a few nonconjugate models.
Fast Function to Function Regression
Junier Oliva,Willie Neiswanger,Barnabas Poczos,Eric Xing,Jeff Schneider
Computer Science , 2014,
Abstract: We analyze the problem of regression when both input covariates and output responses are functions from a nonparametric function class. Function to function regression (FFR) covers a large range of interesting applications including time-series prediction problems, and also more general tasks like studying a mapping between two separate types of distributions. However, previous nonparametric estimators for FFR type problems scale badly computationally with the number of input/output pairs in a data-set. Given the complexity of a mapping between general functions it may be necessary to consider large data-sets in order to achieve a low estimation risk. To address this issue, we develop a novel scalable nonparametric estimator, the Triple-Basis Estimator (3BE), which is capable of operating over datasets with many instances. To the best of our knowledge, the 3BE is the first nonparametric FFR estimator that can scale to massive datasets. We analyze the 3BE's risk and derive an upperbound rate. Furthermore, we show an improvement of several orders of magnitude in terms of prediction speed and a reduction in error over previous estimators in various real-world data-sets.
Asynchronous Parallel Block-Coordinate Frank-Wolfe
Yu-Xiang Wang,Veeranjaneyulu Sadhanala,Wei Dai,Willie Neiswanger,Suvrit Sra,Eric P. Xing
Mathematics , 2014,
Abstract: We develop mini-batched parallel Frank-Wolfe (conditional gradient) methods for smooth convex optimization subject to block-separable constraints. Our work includes the basic (batch) Frank-Wolfe algorithm as well as the recently proposed Block-Coordinate Frank-Wolfe (BCFW) method\citep{lacoste2012block} as special cases. Our algorithm permits asynchronous updates within the minibatch, and is robust to stragglers and faulty worker threads. Our analysis reveals how the potential speedups over BCFW depend on the minibatch size and how one can provably obtain large problem dependent speedups. We present several experiments to indicate empirical behavior of our methods, obtaining significant speedups over competing state-of-the-art (and synchronous) methods on structural SVMs.
Fast Distribution To Real Regression
Junier B. Oliva,Willie Neiswanger,Barnabas Poczos,Jeff Schneider,Eric Xing
Computer Science , 2013,
Abstract: We study the problem of distribution to real-value regression, where one aims to regress a mapping $f$ that takes in a distribution input covariate $P\in \mathcal{I}$ (for a non-parametric family of distributions $\mathcal{I}$) and outputs a real-valued response $Y=f(P) + \epsilon$. This setting was recently studied, and a "Kernel-Kernel" estimator was introduced and shown to have a polynomial rate of convergence. However, evaluating a new prediction with the Kernel-Kernel estimator scales as $\Omega(N)$. This causes the difficult situation where a large amount of data may be necessary for a low estimation risk, but the computation cost of estimation becomes infeasible when the data-set is too large. To this end, we propose the Double-Basis estimator, which looks to alleviate this big data problem in two ways: first, the Double-Basis estimator is shown to have a computation complexity that is independent of the number of of instances $N$ when evaluating new predictions after training; secondly, the Double-Basis estimator is shown to have a fast rate of convergence for a general class of mappings $f\in\mathcal{F}$.
Shocks and Prospects for a Pacific Islands Currency Union  [PDF]
Willie Lahari
Modern Economy (ME) , 2012, DOI: 10.4236/me.2012.35065
Abstract: This analysis re-kindles the debate on the feasibility of a Pacific Islands currency union in view of the recent expansion and consolidation of regional strategies such as the Pacific Plan and the PACER Plus. Limited consideration has been given to the proposition for a Pacific Islands currency union. This paper exploits the OCA theoretical framework and employs the Gonzalo and Ng (2001) decomposition method in investigating the dynamic effects of permanent and transitory shocks on key macroeconomic variables among Pacific Island countries (PICs). Using newly constructed quarterly data in the analysis, evidence shows that the proposed union of six PICs (Fiji, PNG, Samoa, Solomon Islands, Vanuatu and Tonga) do not meet most of the preconditions for a union. However, further investigation shows evidence for the Melanesian countries (Fiji, PNG, Solomon Islands and Vanuatu) to possibly form a monetary union, preferably with the Australian dollar as the anchor currency. Nonetheless, further costs in terms of the alignment of policies by Melanesian countries are required.
Large diffusivity finite-dimensional asymptotic behaviour of a semilinear wave equation
Robert Willie
Journal of Applied Mathematics , 2003, DOI: 10.1155/s1110757x03212067
Abstract: We study the effects of large diffusivity in all parts of the domain in a linearly damped wave equation subject to standard zero Robin-type boundary conditions. In the linear case, we show in a given sense that the asymptotic behaviour of solutions verifies a second-order ordinary differential equation. In the semilinear case, under suitable dissipative assumptions on the nonlinear term, we prove the existence of a global attractor for fixed diffusion and that the limiting attractor for large diffusion is finite dimensional.
NIH names computational boss
Willie Schatz
Genome Biology , 2003, DOI: 10.1186/gb-spotlight-20030605-01
Abstract: Jakobsson replaces James Cassatt, director of the National Institutes of Health's (NIH) NIGMS Division of Cell Biology and Biophysics, who had served as CBCB's acting director since the center was established in 2001. The new director holds dual professorships in the Department of Molecular and Integrative Physiology and in the programs of biophysics, neuroscience, and bioengineering at Urbana-Champaign. He will continue to lead research groups at the university, but will not have teaching responsibilities."It's very good news that NIH finally got someone to move forward with BISTI [the Biomedical Information Science and Technology Initiative]," said Larry Smarr, the founder of the California Institute of Telecommunications and Information Technology. "The BISTI report was presented in June 1999, and very few of the recommendations have been acted on. BISTI is the only road map for the future of NIH-sponsored research for telecommunications and information technology."Implementing BISTI is essential to NIH's developing of the information infrastructure to support a national model for biomedical research, said Smarr, who is Jakobsson's former colleague at the National Center for Supercomputing Applications at the University of Illinois. "There hasn't been any energy or consensus at the agency to implement the BISTI recommendations," Smarr told us. "Hiring Eric shows NIH is ready to do that."The CBCB, which will be the centerpiece of that effort, supports research and training in areas that join biology with the computer sciences, engineering, mathematics, and physics. Such research includes computer modeling of biological networks and dynamic processes and quantitative approaches to cellular, molecular, and developmental biology."The most appealing part of the job is that it's undefined," Jakobsson told us. "I'm envisioning this job as putting together a nationally distributed software engineering project. Now, a program for molecular dynamics and another that simula
Implikasies van die prosesse van transformasie vir die universiteitswese in Suid-Afrika
Willie Esterhuyse
Koers : Bulletin for Christian Scholarship , 1994, DOI: 10.4102/koers.v59i1.654
Abstract: This article addresses a contentious political and policy issue: the implications which the process of change currently taking place in South Africa hold for traditional 'white’ universities. It is argued that this issue should be addressed, on the one hand, from the perspective of sociopolitical transformation and, on the other hand, from the perspective of modernizing development. These two major transformational processes determine the context in which universities need to restructure and position themselves. The implications of these processes for traditional ‘white' universities in general and Afrikaans universities in particular are analysed in terms of three distinctive categories or levels of issues: The nature, function and mission of the university as an academic institution. The organizational structure of the university. The organizational culture, social structure and value system of the university as a community of people. It is concluded that the desegregation of traditional 'white ’ universities has been but the first step in the complicated process of transforming South African universities.
Page 1 /112
Display every page Item

Copyright © 2008-2017 Open Access Library. All rights reserved.