| Bayesian Hierarchical Model Characterization of Model Error in Ocean Data Assimilation and Forecasts |
28 Sep 2012 |
16 pages |
| Authors:
Ralph F Milliff; Christopher K Wikle; L M Berliner; Radu Herbei; NORTHWEST RESEARCH ASSOCIATES BOULDER CO COLORADO RESEARCH ASSOCIATES DIV
|
 | Quantitative uncertainty management attributes of the Bayesian Hierarchical Model (BHM) methodology are applied to the identification, characterization, and modelling of irreducible model error in ocean data assimilation and forecast systems. |
|
| QSAR Classification Model for Antibacterial Compounds and Its Use in Virtual Screening |
26 Sep 2012 |
13 pages |
| Authors:
Narender Singh; Sidhartha Chaudhury; Ruifeng Liu; Mohamed D AbdulHameed; Gregory Tawa; Anders Wallqvist; ARMY MEDICAL RESEARCH AND MATERIEL COMMAND FORT DETRICK MD TELEMEDICINE AND ADVANCED TECH RESEARCH CENTER
|
| As novel and drug-resistant bacterial strains continue to present an emerging health threat, the development of new antibacterial agents is critical. This includes making improvements to existing antibacterial scaffolds as well as identifying novel ones. The aim of this study is to apply a Bayesian classification QSAR approach to rapidly screen chemical libraries for compounds predicted to have antibacterial activity. Toward this end we assembled a data set of 317 ... |
|
| Spectral Approaches to Learning Predictive Representations |
Sep 2012 |
177 pages |
| Authors:
Byron Boots; CARNEGIE-MELLON UNIV PITTSBURGH PA SCHOOL OF COMPUTER SCIENCE
|
| A central problem in artificial intelligence is to choose actions to maximize reward in a partially observable, uncertain environment. To do so, we must obtain an accurate environment model, and then plan to maximize reward. However, for complex domains, specifying a model by hand can be a time-consuming process. This motivates an alternative approach: learning a model directly from observations. Unfortunately, learning algorithms often recover a model that is too ... |
|
| Attitude Estimation for Unresolved Agile Space Objects with Shape Model Uncertainty |
Sep 2012 |
11 pages |
| Authors:
Marcus J Holzinger; Kyle T Alfriend; Charles J Wetterer; K K Luu; Chris Sabol; Kris Hamada; Andrew Harms; AIR FORCE RESEARCH LAB WRIGHT-PATTERSON AFB OH
|
| The problem of estimating attitude for actively maneuvering or passively rotating Space Objects (SOs) with unknown mass properties / external torques and uncertain shape models is addressed. To account for agile SO maneuvers, angular rates are simply assumed to be random inputs (e.g., process noise), and model uncertainty is accounted for in a bias state with dynamics derived using rst principles. Bayesian estimation approaches are used to estimate the resulting ... |
|
| Model Validation for Simulations of Vehicle Systems |
16 Aug 2012 |
24 pages |
| Authors:
Hao Pan; Gregory Hulbert; Michael Kokkolaras; Matthew Castanier; David Lamb; FORD MOTOR CO DEARBORN MI
|
| Outline: Classification of validation approaches, Bayesian interval hypothesis testing, Quantifying model confidence, Distribution-free approach by means of bootstrapping, Statistical power superiority, Validation benchmark problem, ARC-developed electro-thermal battery model validation, for energy & power community of interest application. |
|
| Model Validation for Simulations of Vehicle Systems |
Aug 2012 |
15 pages |
| Authors:
David Lamb; Matthew Castanier; Hao Pan; Michael Kokkolaras; Gregory Hulbert; MICHIGAN UNIV ANN ARBOR DEPT OF MECHANICAL ENGINEERING
|
| This paper deals with model validation of dynamic systems (with vehicle systems being of particular interest) that have multiple time-dependent output. First, we review several validation methodologies that have been reported in the literature: graphical comparison, feature-based techniques, PDF/CDF based techniques, Bayesian posterior estimation, classical hypothesis testing and Bayesian hypothesis testing. We discuss their advantages and disadvantages in terms of several attributes: applicability to different types of models, need for ... |
|
| Statistical Methods for Studying Genetic Variation in Populations |
Aug 2012 |
169 pages |
| Authors:
Suyash Shringarpure; CARNEGIE-MELLON UNIV PITTSBURGH PA MACHINE LEARNING DEPT
|
| The study of genetic variation in populations is of great interest for the study of the evolutionary history of humans and other species. Improvement in sequencing technology has resulted in the availability of many large datasets of genetic data. Computational methods have therefore become quite important in analyzing these data. Two important problems that have been studied using genetic data are population stratification (modeling individual ancestry with respect to ancestral ... |
|
| Statistical Analysis of Hit/Miss Data (Preprint) |
Jul 2012 |
16 pages |
| Authors:
Jeremy S Knopp; Li Zeng; AIR FORCE RESEARCH LAB WRIGHT-PATTERSON AFB OH MATERIALS AND MANUFACTURING DIRECTORATE
|
| There has been renewed interest in the analysis of Hit/Miss or Bernoulli data in the context of Nondestructive Evaluation (NDE). Many contributions have been made with a focus on confidence bound estimation on this type of data. Some concern regarding the proper calculation and use of a90 and a90/95 estimates has been raised. In particular, the behavior of a Probability of Detection (POD) curve for large flaw sizes above stated ... |
|
| Health Assessment and Fault Classification of Roller Element Bearings |
Jul 2012 |
24 pages |
| Authors:
Andrew J Bayba; David N Siegel; Kwok Tom; Derwin Washington; ARMY RESEARCH LAB ADELPHI MD SENSORS AND ELECTRON DEVICES DIRECTORATE
|
| Feature extraction, health assessment, and fault classification algorithms were evaluated for ball bearings with three different fault types and multiple levels of damage. Data was analyzed for five healthy bearings, and seeded fault bearings with five levels of damage for each fault type (ball fault, inner race fault, and outer race fault). A variety of fault analysis techniques were used to calculate properties (features) of the data sets, which were ... |
|
| Framework for Smart Electronic Health Record-Linked Predictive Models to Optimize Care for Complex Digestive Diseases |
Jun 2012 |
11 pages |
| Authors:
Michael A Dunn; Melissa Saul; PITTSBURGH UNIV PA
|
| Our major objective is to develop an electronic application capable of integrating and semantically standardizing electronic medical record (EMR) data to generate de-identified datasets populated with longitudinal clinical data drawn from diverse sources. In Year 1 of our project, we have successfully built the infrastructure to support this project. We have defined and generated the EMR-based datasets to be used for algorithm development. In year 2, we will test the ... |
|
| Surviving in Cyberspace: A Game Theoretic Approach |
Jun 2012 |
16 pages |
| Authors:
Charles A Kamhoua; Kevin A Kwiat; Joon S Park; AIR FORCE RESEARCH LAB ROME NY INFORMATION DIRECTORATE
|
| As information systems become ever more complex and the interdependence of these systems increases, a mission-critical system should have the fight-through ability to sustain damage yet survive with mission assurance in cyberspace. To satisfy this requirement, in this paper we propose a game theoretic approach to binary voting with a weighted majority to aggregate observations among replicated nodes. Nodes are of two types: they either vote truthfully or are malicious ... |
|
| Acquisition Management for System-of-Systems: Requirement Evolution and Acquisition Strategy Planning |
30 Apr 2012 |
54 pages |
| Authors:
Seung Y Han; Zhemei Fang; Daniel DeLaurentis; PURDUE UNIV LAFAYETTE IN SCHOOL OF AERONAUTICS AND ASTRONAUTICS
|
| The complex interdependencies between systems organized for a system-of-systems (SoS) capability pose a challenge to effective acquisition management of SoS assets. In general methodologies to assess risk that cascades through interdependencies are critical to effectively analyzing alternatives in a capability-based acquisition strategy. A particular problem occurs in cases where requirements on systems are evolving. In this paper, a Bayesian Network (BN) method is presented, which models requirement evolution in the ... |
|
| Target Classification of Canonical Scatterers Using Classical Estimation and Dictionary Based Techniques |
22 Mar 2012 |
221 pages |
| Authors:
II Hammond Glenn B; AIR FORCE INST OF TECH WRIGHT-PATTERSON AFB OH GRADUATE SCHOOL OF ENGINEERING AND MANAGEMENT
|
| This research effort will utilize a hierarchical dictionary-based approach for canonical shape classification within measured synthetic aperture radar (SAR) phase history data. This primary goal of this research is to develop an efficient framework for dictionary based SAR feature extraction using modi ed 3-D radar scattering models. Previous work in this area relies on maximum likelihood (ML) estimation and similar approaches to extract shapes using 2-D signal models. We include ... |
|
| Combat Identification of Synthetic Aperture Radar Images Using Contextual Features and Bayesian Belief Networks |
Mar 2012 |
134 pages |
| Authors:
John X Situ; AIR FORCE INST OF TECH WRIGHT-PATTERSON AFB OH GRADUATE SCHOOL OF ENGINEERING AND MANAGEMENT
|
| Given the nearly infinite combination of modifications and configurations for weapon systems, no two targets are ever exactly the same. Synthetic Aperture Radar (SAR) imagery and associated High Range Resolution (HRR) profiles of the same target will have different signatures when viewed from different angles. To overcome this challenge, data from a wide range of aspect and depression angles must be used to train pattern recognition algorithms. Alternatively, features invariant ... |
|
| Modeling of Helicopter Pilot Misperception During Overland Navigation |
Mar 2012 |
110 pages |
| Authors:
Bradley T Cowden; NAVAL POSTGRADUATE SCHOOL MONTEREY CA DEPT OF OPERATIONS RESEARCH
|
| This thesis provides a framework to model human belief and misperception in helicopter overland navigation. Helicopter overland navigation is a challenging mission area because it is a complex cognitive task, and failing to recognize when the aircraft is off-course can lead to operational failures and mishaps. A human-in-the-loop experiment to investigate pilot misperception during simulated overland navigation by analyzing actual navigation trajectory, pilots' perceived location, and corresponding confidence levels was ... |
|
| More Efficient Bayesian-based Optimization and Uncertainty Assessment of Hydrologic Model Parameters |
Feb 2012 |
31 pages |
| Authors:
Brian E Skahill; Jeffrey S Baggett; ENGINEER RESEARCH AND DEVELOPMENT CENTER VICKSBURG MS COASTAL AND HYDRAULICS LAB
|
| An important consideration in assessing the performance of model calibration software is that of run time. Minimizing the number of hydrologic model runs required during the calibration process is nearly always important, but particularly when the objective function landscape contains multiple local minima or hydrologic model run times are high. Minimizing the number of required model runs was one of the primary factors driving the research and development activities encapsulated ... |
|
| Acquiring and Exploiting Rich Causal Models for Robust Decision Making |
01 Jan 2012 |
17 pages |
| Authors:
Joshua B Tenenbaum; Leslie P Kaelbling; Michael L Littman; David Wingate; MASSACHUSETTS INST OF TECH CAMBRIDGE
|
| Our project has made fundamental contributions to the understanding of robust decision making in human beings and machines through an intensive examination of how to learn rich, causal models of the world and how agents can use those models to make decisions. We report progress in eight key areas: 1) Significant progress on building rich models using probabilistic programming. 2) New Bayesian nonparametric models for learning dynamical systems. 3) A ... |
|
| Nonparametric Bayesian Context Learning for Buried Threat Detection |
Jan 2012 |
333 pages |
| Authors:
Christopher R Ratto; DUKE UNIV DURHAM NC DEPT OF ELECTRICAL AND COMPUTER ENGINEERING
|
| This dissertation addresses the problem of detecting buried explosive threats (i.e. landmines and improvised explosive devices) with ground-penetrating radar (GPR) and hyperspectral imaging (HSI) across widely-varying environmental conditions. Automated detection of buried objects with GPR and HSI is particularly difficult due to the sensitivity of sensor phenomenology to variations in local environmental conditions. Past approaches have attempted to mitigate the effects of ambient factors by designing statistical detection and classification ... |
|
| Environmental Fate Model for Ultra-Low-Volume Insecticide Applications Used for Adult Mosquito Management |
Jan 2012 |
10 pages |
| Authors:
Jerome J Schleier III; Robert K Peterson; Kathryn M Irvine; Lucy M Marshall; David K Weaver; Collin J Preftakes; MONTANA STATE UNIV BOZEMAN DEPT OF LAND RESOURCES AND ENVIRONMENTAL SCIENCES
|
| One of the more effective ways of managing high densities of adult mosquitoes that vector human and animal pathogens is ultra-low-volume (ULV) aerosol applications of insecticides. The U.S. Environmental Protection Agency uses models that are not validated for ULV insecticide applications and exposure assumptions to perform their human and ecological risk assessments. Currently, there is no validated model that can accurately predict deposition of insecticides applied using ULV technology for ... |
|
| Dynamic and Supervised Topic Models for Literature-Based Discovery |
21 Dec 2011 |
7 pages |
| Authors:
David M Blei; PRINCETON UNIV NJ DEPT OF COMPUTER SCIENCE
|
| Under the support of the ONR my research focused on extending the state ot the an or probabilistic topic modeling, algorithms for making discoveries from and predictions about large collections of texts. For the past three years, my group has published many papers in the service of this goal. In this report, I will highlight some of the themes and publications that represent this work. Thanks to the support of ... |
|
| Multi-output Local Gaussian Process Regression: Applications to Uncertainty Quantification |
07 Dec 2011 |
63 pages |
| Authors:
Ilias Bilionis; Nicholas Zabaras; CORNELL UNIV ITHACA NY MATERIALS PROCESS DESIGN AND CONTROL LABORATORY (MPDC)
|
| We develop an efficient, Bayesian Uncertainty Quantification framework us- ing a novel treed Gaussian process model. The tree is adaptively constructed using information conveyed by the observed data about the length scales of the underlying process. On each leaf of the tree, we utilize Bayesian Experimental Design techniques in order to learn a multi-output Gaussian process. The constructed surrogate can provide analytical point estimates, as well as error bars, for ... |
|
| Information Selection in Intelligence Processing |
Dec 2011 |
135 pages |
| Authors:
Yuval Nevo; NAVAL POSTGRADUATE SCHOOL MONTEREY CA DEPT OF OPERATIONS RESEARCH
|
| In many intelligence agencies, the processing of data into usable information ready for analysis poses a significant bottleneck. Typically, much more data is available than can be processed in the limited time available for processing. We formulate the problem faced by an intelligence collection unit when processing incoming raw information for delivery to intelligence analyst as an exploration-exploitation problem: the processor has to choose between exploring for new sources of ... |
|
| Model-Based Structural Health Monitoring of Fatigue Damage Test-Bed Specimens |
15 Nov 2011 |
26 pages |
| Authors:
Kincho H Law; Jerome P Lynch; Masahiro Kurata; MICHIGAN UNIV ANN ARBOR DEPT OF ELECTRICAL ENGINEERING AND COMPUTER SCIENCE
|
| This research project represents a preliminary investigation into a comprehensive analytical, numerical, and experimental framework for the monitoring and life-cycle assessment of the structural integrity and performance of aluminum hull structures. Aluminum plate specimens were designed and fabricated to facilitate the investigation of system identification and damage detection methodologies, both of which are key components of future life-cycle analyses. The design of the plate specimens was intended to include the ... |
|
| Probabilistic Collocation Method for NDE Problems with Uncertain Parameters with Arbitrary Distributions (Preprint) |
Nov 2011 |
10 pages |
| Authors:
Jeremy S Knopp; Mark P Blodgett; M R Cherry; AIR FORCE RESEARCH LAB WRIGHT-PATTERSON AFB OH MATERIALS AND MANUFACTURING DIRECTORATE
|
| In order to quantify the reliability of NDE systems, large amounts of experiments are performed to develop a probability of detection (POD) curve for the system. These POD studies require a substantial amount of experimentation which can sometimes be cost prohibitive. To expedite the process of developing these curves, highly precise numerical models are used in conjunction with NDE sensors to understand the uncertainties associated with the inspections. Numerical models ... |
|
| Trust, Opinion Diffusion and Radicalization in Social Networks |
Nov 2011 |
6 pages |
| Authors:
Lin Li; Anna Scaglione; Ananthram Swami; Qing Zhao; CALIFORNIA UNIV DAVIS DEPT OF ELECTRICAL AND COMPUTER ENGINEERING
|
| Gossiping models have increasingly been applied to study social network phenomena. This paper is specifically concerned with modeling how the opinions of social agents can be radicalized if the agents interact more strongly with neighbors that share their beliefs. In our model, each agent's belief is represented by a vector of probabilities that a given state is true. The agents average their opinions with that of their neighbors over time, ... |
|
| Adaptive Hessian-based Non-stationary Gaussian Process Response Surface Method for Probability Density Approximation with Application to Bayesian Solution of Large-scale Inverse Problems |
Oct 2011 |
38 pages |
| Authors:
Tan Bui-Thanh; Omar Ghattas; David Higdon; TEXAS UNIV AT AUSTIN INST FOR COMPUTATIONAL ENGINEERING AND SCIENCES
|
| We develop an adaptive Hessian-based non-stationary Gaussian process response surface method to approximate a probability density function (pdf) that exploits its structure, in particular the Hessian of its negative logarithm. Of particular interest to us are pdfs that arise from the Bayesian solution of large-scale inverse problems, which imply very expensive-to-evaluate pdfs. The method can be considered as a piecewise adaptive Gaussian approximation in which a Gaussian tailored to the ... |
|
| Predicting Tropical Cyclone Genesis |
30 Sep 2011 |
7 pages |
| Authors:
Melinda S Peng; James Hansen; Tim Li; NAVAL RESEARCH LAB MONTEREY CA
|
| The long-term goal of this research is to provide probabilistic genesis forecast guidance to operational forecasters and develop a genesis index for operational dynamical model prediction of tropical cyclone (TC) genesis. Once regions of high TC genesis probability are identified, a movable, multi-nested version of COAMPS with resolution of roughly 3 km or less in the inner most grid will be used for predicting the genesis event. The objective of ... |
|
| Efficient Inversion in Underwater Acoustics with Iterative and Sequential Bayesian Methods |
30 Sep 2011 |
10 pages |
| Authors:
Zoi-Heleni Michalopoulou; NEW JERSEY INST OF TECH NEWARK DEPT OF MATHEMATICAL SCIENCES
|
| LONG TERM GOALS: The long term goal of this project is to develop efficient inversion algorithms for successful geoacoustic parameter estimation and source localization, exploiting (fully or partially) the physics of the propagation medium. Algorithms are designed for geoacoustic inversion via the extraction features of the acoustic field. OBJECTIVES: Achieve accurate and computationally efficient geoacoustic inversion and source localization by designing estimation schemes that combine acoustic field modeling and statistical ... |
|
| A Sustainable Approach for Optimal Steel Sheet Pile Structure Assessment, Maintenance, and Rehabilitation |
30 Sep 2011 |
108 pages |
| Authors:
Kevin L Rens; Rui Liu; Stuart Foltz; ENGINEER RESEARCH AND DEVELOPMENT CENTER CHAMPAIGN IL CONSTRUCTION ENGINEERING RESEARCH LAB
|
| The U.S. Army Corps of Engineers (USACE) has constructed a wide variety of civil works structures. Due to age and other factors, many of these structures have deteriorated to a point that they need varying levels of maintenance and repair (M&R). Steel sheet pile (SSP) structures are part of the USACE civilian projects such as lock and dam and other navigation facilities. Failure of a SSP wall or cell can ... |
|
| Mining and Querying Multimedia Data |
29 Sep 2011 |
144 pages |
| Authors:
Fan Guo; CARNEGIE-MELLON UNIV PITTSBURGH PA SCHOOL OF COMPUTER SCIENCE
|
| The emerging popularity of multimedia data, as digital representation of text image, video and countless other milieus, with prodigious volumes and wild diversity exhibits the phenomenal impact of modern technologies in reforming the way information is accessed, disseminated, digested and retained. This has iteratively ignited the data-driven perspective of research and development, to characterize perspicuous patterns, crystallize informative insights, and realize elevated experience for end-users, where innovations in a spectrum ... |
|
| Combining Imaging and Non-Imaging Observations for Improved Space-Object Identification |
27 Sep 2011 |
27 pages |
| Authors:
Sudhakar Prasad; David Brady; Robert Plemmons; NEW MEXICO UNIV ALBUQUERQUE
|
| The accomplishments may be subdivided according to the project s theoretical, experimental, and post-processing/computational objectives. Among the theoretical accomplishments, we list a new model for a sparse representation of man-made space objects and its use, via a new spectral-correlation approach, to segment their material components; the use of 2D segment-boundary data for multiple poses of a man-made space object to recover its 3D shape; new fundamental results involving statistical information ... |
|
| Mathematical Modeling to Reduce the Cost of Complex System Testing: Characterizing Test Coverage to Assess and Improve Information Return |
21 Sep 2011 |
88 pages |
| Authors:
Karl D Pfeiffer; Valery A Kanevsky; Thomas J Housel; NAVAL POSTGRADUATE SCHOOL MONTEREY CA GRADUATE SCHOOL OF OPERATIONAL AND INFORMATION SCIENCES
|
| Effective, cost-efficient testing is critical to the long-term success of Open Architecture within the Navy's Integrated Warfare System. In previous research, we developed a simple, effective framework for examining the testing of complex systems. This model and its prototype decision aid provide a rigorous yet tractable approach to improve system testing, and to better understand and document the system and component interdependencies across the enterprise. An integral part of this ... |
|
| Mine Burial Expert System for Change of MIW Doctrine |
Sep 2011 |
160 pages |
| Authors:
Christopher M Beuligmann; NAVAL POSTGRADUATE SCHOOL MONTEREY CA
|
| Mine impact burial models such as IMPACT25, IMPACT28, and IMPACT35 have been used in the MIW community in an attempt to calculate the percentage of impact burial for sea mines. Until recently the models have been deterministic, using parameters such as sediment type, air and sea trajectories, drop angle, and mine type to calculate the percentage of burial. These models have been relatively effective in calculating impact burial, but little ... |
|
| Bayesian Treaty Monitoring: Preliminary Report |
Sep 2011 |
12 pages |
| Authors:
Stuart J Russell; Stephen C Myers; Nimar S Arora; David A Moore; Erik Sudderth; CALIFORNIA UNIV BERKELEY
|
| Our project has initiated and will develop and evaluate a new Bayesian approach for nuclear test monitoring. We anticipate that the new approach will yield substantially lower detection thresholds, possibly approaching a theoretical lower bound that we hope to establish. We will also develop new techniques to implement such monitoring capabilities within a general-purpose Bayesian modeling and inference system that may eventually support a wide range of information-system needs for ... |
|
| Combining Analyst and Waveform-Correlation-Based Arrival Time Measurements in the Bayesloc Multiple-Event Location Algorithm |
Sep 2011 |
8 pages |
| Authors:
Stephen C Myers; Gardar Johannesson; Douglas A Dodge; Nathan A Simmons; LAWRENCE LIVERMORE NATIONAL LAB CA
|
| We add relative arrival-time measurements that are derived from waveform correlation to the Bayesloc multiple-event location algorithm. Bayesloc is a formulation of joint probability over event locations, travel time corrections phase labels, and arrival-time measurement errors. The Bayesloc formulation is hierarchical with distinct statistical models for each component of the multiple-event system, including prior constraints for any of the parameters. Bayes' Theorem allows calculation of the joint probability for hypothesized ... |
|
| Quantifying Geoacoustic Uncertainty and Seabed Variability for Propagation Uncertainty |
Sep 2011 |
12 pages |
| Authors:
Jan Dettmer; Stan E Dosso; Charles W Holland; VICTORIA UNIV (BRITISH COLUMBIA) SCHOOL OF EARTH AND OCEAN SCIENCES
|
| LONG-TERM GOALS: Propagation and reverberation of acoustic fields in shallow waters depend strongly on the spatial variability of seabed geoacoustic parameters, and lack of knowledge of seabed variability is often a limiting factor in acoustic modeling applications. However, direct sampling (e.g., coring) of vertical and lateral variability is expensive and laborious, and matched-field and other long-range inversion methods fail to provide sufficient resolution. The long-term goal of this work is ... |
|
| Supervised Learning in CINets |
Jul 2011 |
9 pages |
| Authors:
Paul Bruhn; Jeffrey Weinschenk; PENNSYLVANIA STATE UNIV STATE COLLEGE APPLIED RESEARCH LAB
|
| Continuous Inference Networks (CINets), a form of multilayer fuzzy value networks allow computation with fuzzy values in concise structures, are capable of universal function approximation, and are readily interpretable through natural language, aiding maintenance modification, collaboration, and knowledge sharing. However CINets have been reliant on Subject Matter Expertise (SME) and manual tuning to realize optimal performance, limiting their applicability. With ONR support[i], ARL has developed a supervised learning process for ... |
|
| Bayesian Methods for Image Segmentation (Preprint) |
Jun 2011 |
6 pages |
| Authors:
Mary Comer; Charles A Bouman; Marc de Graef; Jeff P Simmons; AIR FORCE RESEARCH LAB WRIGHT-PATTERSON AFB OH MATERIALS AND MANUFACTURING DIR METALS CERAMICS AND NONDESTRUCTIVE EVALUATION DIV/METALS BRANCH
|
| This paper presents an introduction to Bayesian methods for image segmentation, and provides some examples of the performance of these methods. Bayesian image segmentation methods represent a class of statistical approaches to the problem of segmentation. The idea behind Bayesian techniques is to use statistical image models to incorporate prior information into the segmentation process. This is typically done by specifying a model for the observed image to be segmented ... |
|
| Extending Bayesian Logic Programs for Plan Recognition and Machine Reading |
MAY 2011 |
38 pages |
| Authors:
Sindhu V. Raghavan; TEXAS UNIV AT ARLINGTON DEPT OF COMPUTER SCIENCE
|
| Statistical relational learning (SRL) is the area of machine learning that integrates both first-order logic and probabilistic graphical models. The advantage of these formalisms is that they can handle both uncertainty and structured/relational data. As a result, they are widely used in domains like social network analysis, biological data analysis, and natural language processing. Bayesian Logic Programs (BLPs), which integrate both first-order logic and Bayesian networks are a powerful SRL ... |
|
| Bayes-Optimal Methods for Simulation Optimization |
18 Apr 2011 |
46 pages |
| Authors:
Peter I Frazier; CORNELL UNIV ITHACA NY SCHOOL OF OPERATIONS RESEARCH AND INFORMATION ENGINEERING
|
|
| Towards a Comprehensive Framework for Simulation-based Vehicle Systems Design Validation (PREPRINT) |
01 MAR 2011 |
15 pages |
| Authors:
M. Kakkolaras; G. Hulbert; P. Papalambros; Z. Mourelatos; R. J. Yang; M. Brudnak; D. Gorsich; MICHIGAN UNIV ANN ARBOR DEPT OF MECHANICAL ENGINEERING
|
| We present an overview of our most recent and ongoing research efforts to develop a comprehensive framework for simulation-based vehicle design validation. Specifically we present the three major building blocks of our framework, namely i) the investigation of existing and introduction of appropriate validation metrics for comparing the dynamic responses of vehicle systems that consist of multivariate functional data, ii) the selection and robust implementation of a Bayesian interval-based hypothesis ... |
|
| Multidisciplinary Design Optimization Under Uncertainty: An Information Model Approach (PREPRINT) |
01 MAR 2011 |
39 pages |
| Authors:
James A. Reneke; Margaret M. Wiecek; Georges M. Fadel; Sundeep Samson; CLEMSON UNIV SC DEPT OF MATHEMATICAL SCIENCES
|
| Motivated by needs of concurrent multi-disciplinary design of a multi-purpose vehicle, a modeling and methodological approach to handling tradeoffs is presented. Each component has uncertain elements and a random performance which is influenced by the performance of other components. The components may require different knowledge bases and models with different mathematical structures, time and size scales, calling for higher-level coordination. The theory of reproducing kernel Hilbert spaces provides the mathematical ... |
|
| Adaptive Information Filtering |
25 Feb 2011 |
6 pages |
| Authors:
Yi Zhang; CALIFORNIA UNIV SANTA CRUZ
|
| This project studies personalized proactive information filtering agents that pushes relevant information to the user without requiring explicit user query. To do this, the agent adaptively learns a detailed user model while observing and interacting with the user. We use Bayesian statistical theory and machine learning techniques to tackle the following two major challenges. We studied two major problems: how to build an initial user profile with minimal user effort, ... |
|
| Improving Environmental Model Calibration and Prediction |
18 JAN 2011 |
68 pages |
| Authors:
Jeffrey S. Baggett; WISCONSIN UNIV-LA CROSSE
|
| First, we have continued to develop tools for efficient global optimization of environmental models. Our algorithms are hybrid algorithms that combine evolutionary strategies for global search with derivative-free methods for local search in novel and efficient ways. Results thus far are promising and are pointing the way toward practical hybrid optimization tools for environmental models. Second, we are applying function approximation techniques to improve the efficiency of Monte Carlo Markov ... |
|
| Probabilistic Graphical Models for the Analysis and Synthesis of Musical Audio |
Nov 2010 |
126 pages |
| Authors:
Matthew D Hoffman; PRINCETON UNIV NJ
|
| Content-based Music Information Retrieval (MIR) systems seek to automatically extract meaningful information from musical audio signals. This thesis applies new and existing generative probabilistic models to several content-based MIR tasks: timbral similarity estimation, semantic annotation and retrieval, and latent source discovery and separation. In order to estimate how similar two songs sound to one another, we employ a Hierarchical Dirichlet Process (HDP) mixture model to discover a shared representation of ... |
|
| Particle Kalman Filtering for Ocean State Estimation |
30 SEP 2010 |
7 pages |
| Authors:
Ibrahim Hoteit; Aneesh Subramanian; Bruce Cornuelle; SCRIPPS INSTITUTION OF OCEANOGRAPHY LA JOLLA CA
|
| New nonlinear filtering algorithms were developed and are currently being tested. Numerical results suggest that nonlinear filters behave better than the ensemble Kalman filter methods with strongly nonlinear systems. They also seem to respect the dynamical of the system state more resulting in more stable predictions. |
|
| Human Predictive Reasoning for Group Interactions |
SEP 2010 |
49 pages |
| Authors:
Marcus B. Perry; Patrick J. Vincent; Jeremy D. Jordan; ALABAMA UNIV UNIVERSITY
|
| In this effort, two approaches to predicting attitude and behavior of human groups are developed. One approach employs a novel statistical model for assessing group-level and individual-level factors on group constructs of interest. The second approach is more probabilistic in nature, and involves the use of a Markov chain with Bayesian updates in order to capture changes in group constructs as a function of changes in important environmental variables. These ... |
|
| Multiple-Array Detection, Association and Location of Infrasound and Seismo-Acoustic Events - Utilization of Ground Truth Information |
Sep 2010 |
13 pages |
| Authors:
Stephen J Arrowsmith; Il-Young Che; Christopher T Hayward; Brian W Stump; LOS ALAMOS NATIONAL LAB NM
|
| This work is intended to provide automated methodology for processing seismic and infrasound data from seismo-acoustic arrays and apply the methodology to regional networks for validation with ground truth information. In the initial year of the project automated techniques for detecting, associating and locating infrasound signals were developed. Recently, the location procedure has been cast into a Bayesian framework. Work this year has focused on assessment of sets of robust ... |
|
| Probability Based Integration of Structural Health Monitoring into the Aging Aircraft Sustainment Program |
02 AUG 2010 |
9 pages |
| Authors:
Raphael T. Haftka; Fuh-Gwo Yuan; Nam-Ho Kim; FLORIDA UNIV GAINESVILLE
|
| The research focused on improvements in diagnosis and prognosis of crack detection through extensive use of probabilistic techniques. A unique feature of the research is that it identifies the material properties relevant to damage propagation at the same time that it performs diagnosis and prognosis. As such, it has the potential of turning aircraft into flying fatigue laboratories and contributing to substantial improvements in the accuracy of aircraft digital twins. ... |
|
| Theory-based Bayesian Models of Inductive Inference |
19 Jul 2010 |
11 pages |
| Authors:
Thomas L Griffiths; CALIFORNIA UNIV BERKELEY SPONSORED PROJECTS OFFICE
|
| The proposed research had the goal of developing computational models of human inferences that bridge the gap between human and machine learning. This research focused on two components of cognition: learning about categories and their properties, and learning about causal and social relations. Research was completed successfully in both of these areas, resulting in a new unifying framework for models of category learning, new models for how people form and ... |
|
Reports by Keyword(s)
BAYES THEOREM
