Eliciting information from experts for use in constructing prior distributions for logistic regression coefficients can be challenging. The task is especially difficult when the model contains many predictor variables, because the expert is asked to provide summary information about the probability of “success” for many subgroups of the population. Often,...
This paper examines the regulatory decision problem facing fishery managers tasked with balancing desirable fish catch with undesirable bycatch. I approach the problem using Bayesian decision analysis. I first estimate a model of production, accounting for the fishery’s joint production of desirable and undesirable outputs. I then use the estimated...
This study investigated variation in xylem anatomy, hydraulic properties, and the relationship between anatomy and properties within Douglas-fir trees at multiple scales. The hierarchical scales in the study included fertilization treatments (fertilized and unfertilized), trees within the treatments, and positions within the trees. Tracheid diameter, tracheid length, percent latewood, number...
This study investigated variation in xylem anatomy, hydraulic properties, and the relationship between anatomy and properties within Douglas-fir trees at multiple scales. The hierarchical scales in the study included fertilization treatments (fertilized and unfertilized), trees within the treatments, and positions within the trees. Tracheid diameter, tracheid length, percent latewood, number...
A dynamic approach to the Bayesian theory of sampling inspection
by attributes for the single sampling case is presented. A model is
developed which assumes a sequence of lots of equal known size and
lots generated by a process operating in a random manner. The model
also assumes constant associated...
Motivation: The goal of any parentage analysis is to identify as many parent-offspring relationships as possible, while minimizing incorrect assignments. Existing methods can achieve these ends, but require additional information in the form of demographic data, thousands of markers, and/or estimates of genotyping error rates. For many non-model systems, it...
A Bayesian approach to the analysis of a two-phase linear
regression model is given. It is assumed that the regression model is
continuous at the change point. The likelihood function is expressed
in a form which explicitly contains the continuity restriction. The
natural conjugate prior distribution for the likelihood function...
This dissertation explores and analyzes the performance of several Bayesian anytime inference algorithms for dynamic influence diagrams. These algorithms are compared on the On-Line Maintenance Agent testbed, a software artifact permitting comparison of dynamic reasoning algorithms used by an agent on a variety of simulated maintenance and monitoring tasks. Analysis...
Bayesian inferential methods for the two parameter Weibull (and
extreme-value distribution) are presented in a life-testing context. A
practical method of calculating posterior distributions of the two parameters
and a large class of functions of the parameters is presented.
The emphasis is for the situation where the sample information is...
Over the past decade, it has come to light that many published scientific findings cannot be reproduced. This has led to the replication crisis in science. Many researchers feel that they can no longer trust much of what they read in scientific journals, and the public is becoming ever more...
Probabilistic inference using Bayesian networks is now a well-established approach for reasoning under uncertainty. Among many e ciency-driven tech- niques which have been developed, the Optimal Factoring Problem (OFP) is distinguished for presenting a combinatorial optimization point of view on the problem. The contribution of this thesis is to extend...
An important challenge in machine learning is to find ways of learning quickly from very small amounts of training data. The only way to learn from small data samples is to constrain the learning process by exploiting background knowledge. In this report, we present a theoretical analysis on the use...
Bayesian Optimization (BO) methods are often used to optimize an unknown function f(•) that is costly to evaluate. They typically work in an iterative manner. In each iteration, given a set of observation points, BO algorithms select k ≥ 1 points to be evaluated. The results of those points are...
Finite order autoregressive models for time series are often
used for prediction and other inferences. Given the order of the
model, the parameters of the models can be estimated by least
squares, maximum likelihood, or the Yule-Walker method. The
basic problem is estimating the order of the model. A number...
Image segmentation continues to be a fundamental problem in computer vision and image understanding. In this thesis, we present a Bayesian network that we use for object boundary detection in which the MPE (most probable explanation) before any evidence can produce multiple non-overlapping, non-self-intersecting closed contours and the MPE with...
Though long a standard technique in engineering and medical research, duration or survival analysis has
become common in economics only in recent decades, and in fisheries economics we are aware of only
one previous study. In this paper, we demonstrate the usefulness of duration analysis to understanding
fleet dynamics, specifically...
Importance sampling algorithms are discussed in detail, with an emphasis on implicit sampling, and applied to data assimilation via particle filters. Implicit sampling makes it possible to use the data to find high-probability samples at relatively low cost, making the assimilation more efficient. A new analysis of the feasibility of...
In this paper, we review our experiences in combining cross disciplinary, probabilistic information by Bayesian networks. They seem to be potential tools in combining the optimization orientated economic
modeling tradition to the somewhat different traditions of sociological sciences and further to the empirically orientated biological approaches. Calculus is based on...
The marine ecosystem provides a broad spectrum of services for a wide array of user groups. Such broad use of marine resources often results in use conflicts between ecosystem elements and human interests. To help mitigate these conflicts, the Bayesian Analysis for Spatial Siting (BASS) tool was developed to incorporate...
Bayesian networks are probabilistic models that represent complex distributions in a modular way and have become very popular in many fields. There are many methods to build Bayesian networks from a random sample of independent and identically distributed observations. However, many observational studies are designed using some form of clustered...
Machine learning encompasses probabilistic and statistical techniques that can build models from large quantities of extensional information (examples) with minimal dependence on intensional information (domain knowledge). This focus of machine learning is reflected in the never-ending quest for "off-the-shelf" classifiers. To generalize to unseen data, however, we must make use...
Debanoir is a distributed, collaborative editor for building Bayesian networks. It is composed of a front-end editor client and back-end server programs. The editor client is a Java applet that can be invoked from any Internet browser that supports Java. The server, as of now, runs on the machine from...
For the west coast of North America, from northern California to southern Washington, a habitat suitability prediction framework was developed to support wave energy device siting. Concern that wave energy devices may impact the seafloor and benthos has renewed research interest in the distribution of marine benthic invertebrates and factors...
Bayesian optimization (BO) aims to optimize costly-to-evaluate functions by running a limited number of experiments that each evaluates the function at a selected input. Typical BO formulations assume that experiments are selected sequentially, or in fixed batches. Moreover, these experiments can be executed immediately upon request and have the same...
A copula is the representation of a multivariate distribution. Copulas are used to model multivariate data in many fields. Recent developments include copula models for spatial data and for discrete marginals. We will present a new methodological approach for modeling discrete spatial processes and for predicting the process at unobserved...
Graphical models use Markov properties to establish associations among dependent variables. To estimate spatial correlation and other parameters in graphical models, the conditional independences and joint probability distribution of the graph need to be specified. We can rely on Gaussian multivariate models to derive the joint distribution when all the...
This thesis describes the application of Bayesian networks for monitoring and
diagnosis of a multi-stage manufacturing process, specifically a high speed production
part at Hewlett Packard. Bayesian network "part models" were designed to represent
individual parts in-process. These were combined to form a "process model", which is a
Bayesian network...
In Bayesian analysis, means are commonly used to
summarize Bayesian posterior distributions. Problems with
a large number of parameters often require numerical
integrations over many dimensions to obtain means. In this
dissertation, posterior modes with respect to appropriate
measures are used to summarize Bayesian posterior
distributions, using the Newton-Raphson method...
Commodity equivalence and population similarity are two widely accepted paradigms for the
valid transfer of welfare estimates across resource valuation contexts. We argue that strict
adherence to these rules may leave relevant information untapped. We propose a Bayesian
model search algorithm that examines the probabilities with which two or more...
There are a nearly unlimited number of situations in which the
status of time-varying processes must be updated. The monitoring of
these processes usually occurs at periodic intervals. Whether the
monitoring is performed by man or machine, a decision must be made
regarding the frequency of these activities, that is,...
The paper puts forward a Bayesian Network model to study the optimal eutrophication control in coastal waters by reducing nutrient loads and removing fish biomass (bottom-up and top-down ecosystem mechanisms). The model combines an aquatic ecosystem model with an economic model and examines the economic and ecological consequences of nutrient...
How can an agent generalize its knowledge to new circumstances? To learn
effectively an agent acting in a sequential decision problem must make intelligent action selection choices based on its available knowledge. This dissertation focuses on Bayesian methods of representing learned knowledge and develops novel algorithms that exploit the represented...
Published May 1977. Facts and recommendations in this publication may no longer be valid. Please look for up-to-date information in the OSU Extension Catalog: http://extension.oregonstate.edu/catalog
This work – in which three peer-reviewed academic papers are presented – addresses the ap-plication of Bayesian Reinforcement Learning to the control of a class of ocean wave energy conversion systems. The first paper presents a comparison of a Reinforcement-Learning (RL) based wave energy converter controller against standard Reactive Damping...
Re-establishing connectivity is a primary restoration activity for enhancing the
recovery of migratory fishes, but actions are often limited by lack of funds and
understanding of the benefits of individual projects. The objective of this study was to
develop a Bayesian Network (BN) to assess priorities for restoration of aquatic...
While much work has been done in estimating software reliability, little attention is paid to predict reliability as early as at the design time. In this report, we present our initial research results of building an early stage software reliability prediction model.
In Part I, we will first investigate and...
This thesis describes research to implement a Bayesian belief network based
expert system to solve a real-world diagnostic problem troubleshooting integrated
circuit (IC) testing machines. Several models of the IC tester diagnostic problem
were developed in belief networks, and one of these models was implemented
using Symbolic Probabilistic Inference (SPI)....
Logic Sampling, Likelihood Weighting and AIS-BN are three variants of stochastic sampling, one class of approximate inference for Bayesian networks. We summarize the ideas underlying each algorithm and the relationship among them. The results from a set of empirical experiments comparing Logic Sampling, Likelihood Weighting and AIS-BN are presented. We...
The problem of handling dependent evidence is an important practical issue for applications of reasoning with uncertainty in artificial intelligence. The existing solutions to the problem are not satisfactory because of their ad hoc nature, complexities, or limitations. In this dissertation, we develop a general framework that can be used...
The discovery of GW170817 provided the first empirical evidence that merging binary neutron star systems are both progenitors of short gamma-ray bursts, as well as the primary sites of the nucleosynthetic rapid-neutron capture process. Initially detected as gravitational wave (GW) and gamma-ray burst (GRB) triggers, GW170817 was well-localized and follow-up...
Hewlett-Packard (HP) is one of the world's largest computer companies and the foremost producer of test and measurement instruments. In Corvallis, Oregon, HP manufactures several precision products on high speed, automated assembly lines. The alignment process of a cap to a base part is one of the essential processes in...
Knowledge about the relationship between habitat structure and abundance of a target species
facilitates biodiversity conservation in managed forests. However, modeling the relationship
for infrequent small mammal species in silvicultural experiments introduces the challenge of
excessive zero counts and complex hierarchical sampling. A common solution has been to
ignore infrequent...
Mitigating for increased human impact to the seafloor associated with resource extraction activities and renewable energy development can benefit from an understanding of the distribution of sensitive marine benthic species. Habitat suitability predictive modeling is a cost effective statistical tool to infer species distribution patterns from constrained sampling locations. However,...
Analyzing systems during the conceptual stages of design for characteristics essential to the ease of fault diagnosis is important in today's mechanical systems because consumers and manufacturers are becoming increasingly concerned with cost incurred over the life cycle of the system. The increase in complexity of modem mechanical systems can...